think aloud research

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• My Account Login
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 30 September 2021

The think aloud paradigm reveals differences in the content, dynamics and conceptual scope of resting state thought in trait brooding

• Quentin Raffaelli   ORCID: orcid.org/0000-0003-4046-9818 1 ,
• Caitlin Mills   ORCID: orcid.org/0000-0003-4498-0496 2 ,
• Nadia-Anais de Stefano 1 ,
• Matthias R. Mehl   ORCID: orcid.org/0000-0003-2698-5007 1 ,
• Kate Chambers 1 ,
• Surya A. Fitzgerald 1 ,
• Ramsey Wilcox 1 ,
• Kalina Christoff   ORCID: orcid.org/0000-0003-2841-8647 3 , 4 ,
• Eric S. Andrews   ORCID: orcid.org/0000-0001-7186-5755 1 ,
• Matthew D. Grilli   ORCID: orcid.org/0000-0003-0089-3243 1 , 5 , 6 ,
• Mary-Frances O’Connor   ORCID: orcid.org/0000-0001-5961-6350 1 &
• Jessica R. Andrews-Hanna   ORCID: orcid.org/0000-0003-2565-4759 1 , 6 , 7  

Scientific Reports volume  11 , Article number:  19362 ( 2021 ) Cite this article

5945 Accesses

11 Citations

137 Altmetric

Metrics details

• Human behaviour

Although central to well-being, functional and dysfunctional thoughts arise and unfold over time in ways that remain poorly understood. To shed light on these mechanisms, we adapted a “think aloud” paradigm to quantify the content and dynamics of individuals’ thoughts at rest. Across two studies, external raters hand coded the content of each thought and computed dynamic metrics spanning duration, transition probabilities between affective states, and conceptual similarity over time. Study 1 highlighted the paradigm’s high ecological validity and revealed a narrowing of conceptual scope following more negative content. Study 2 replicated Study 1’s findings and examined individual difference predictors of trait brooding, a maladaptive form of rumination. Across individuals, increased trait brooding was linked to thoughts rated as more negative, past-oriented and self-focused. Longer negative and shorter positive thoughts were also apparent as brooding increased, as well as a tendency to shift away from positive conceptual states, and a stronger narrowing of conceptual scope following negative thoughts. Importantly, content and dynamics explained independent variance, accounting for a third of the variance in brooding. These results uncover a real-time cognitive signature of rumination and highlight the predictive and ecological validity of the think aloud paradigm applied to resting state cognition.

Similar content being viewed by others

Dreams share phenomenological similarities with task-unrelated thoughts and relate to variation in trait rumination and COVID-19 concern

Brain meta-state transitions demarcate thoughts across task contexts exposing the mental noise of trait neuroticism

Repetitive negative thinking in daily life and functional connectivity among default mode, fronto-parietal, and salience networks

Introduction.

Contemplated upon by early Buddhists and later by William James, the stream of thought, or consciousness, is an obscure but ubiquitous aspect of our waking life 1 . Although dynamics are a central feature of conscious experience, static methodologies such as retrospective recall and experience sampling have dominated our mainstream understanding of human thought. As a result, relatively little is known about the way thoughts arise and unfold over time, especially during idle time when left to our own musings 2 . The dynamics of thought may offer unique insight into the relationship between mental life and mental health. Indeed, although repetitive thinking may be beneficial at times 3 , 4 , rumination and worry are both characterized by maladaptive repetitive thinking , a transdiagnostic construct fueling maladaptive outcomes across multiple diagnostic categories 5 , 6 . Rumination in particular (especially the brooding facet) may signify a maladaptive form of coping involving negative mood with detrimental effects on the onset, length, and severity of depressive symptoms 7 , 8 , 9 .

According to our Dynamic Neurocognitive Framework, rumination can be considered a form of non-deliberate thinking guided by high “automatic” constraints that limit the conceptual scope and dynamics of thought 10 , 11 , 12 . Automatically constrained thoughts may feel “sticky” because they are pulled by salient affective or sensory information and, as a result, may be difficult to disengage from. An important feature of automatic constraints is their potential to differentiate rumination from spontaneous forms of thinking 13 -such as mind-wandering and dreaming—that move freely over time without constraints to guide them 11 , 12 , 14 , 15 . Whereas ruminative thinking may worsen mood at subsequent timepoints 16 , 17 , freely-moving thoughts are associated with positive subsequent mood 18 .

Despite theories highlighting the repetitive and inflexible nature of automatically-constrained cognition, relatively little empirical work has quantified the dynamic trajectories of naturally-occurring thoughts in ruminative and non-ruminative individuals. Sticky thoughts could manifest as thoughts of longer duration, as shorter thoughts that resurface more frequently over time, or both. How the affective content of thought impinges upon its duration and recurrence also remains to be delineated in both ruminative and non-ruminative individuals.

Of relevance is the concept of emotional inertia , the propensity to maintain one’s affective state from one moment to the next 19 . Emotional inertia is conceptualized as a maladaptive lack of emotional reactivity to the environment, where one’s current emotional state is maintained despite situational changes in which a different emotional response would be more adaptive 19 . Interestingly, higher trait rumination tends to predict stronger emotional inertia for both negative and positive emotions 20 , though negative emotional inertia is stronger in depression 20 , 21 . Although emotional inertia characterizes affective as opposed to cognitive inflexibility, emotional inertia theories may be relevant for conscious thought because individuals who tend to get “stuck” affectively may be more likely to get stuck cognitively.

Other theories, including the Broaden-and-Build theory of Fredrickson and colleagues 22 , 23 , suggest that positive and negative affect may have opposite effects on cognitive flexibility, with the former sparking more flexible and exploratory forms of cognition, and the latter leading to more rigid forms of cognition. This tendency may be self-perpetuating, creating upward or downward spirals of cognitive-affective states 23 which may be exacerbated in ruminative individuals. In their Attentional Scope Model of Rumination, Whitmer and Gotlib 24 propose that the array of thoughts, percepts, and actions may be inherently narrower in individuals who ruminate more frequently, regardless of whether ruminative individuals are experiencing positive or negative moods. This narrowing of attention aligns with findings that individuals with high trait rumination can exhibit stronger maintenance of task-relevant goals, leading to better performance on executive function tasks emphasizing cognitive stability 25 . Understanding how trait rumination alters the dynamics and conceptual scope of thought is a major goal of the current study.

Notably, the clinical relevance of cognitive rigidity may be amplified when individuals are left to the mercy of their thoughts for extended periods of time, especially without an external task or activity to distract themselves. Indeed, in early sensory deprivation studies, some participants reported difficulty being alone with their thoughts and began to “mind wander in a disturbing fashion” 26 , 27 , findings convergent with a recent empirical study 28 . In more typical contexts, excessive down time for ruminative individuals may allow automatic modes of thinking to resurface, which may exacerbate negative affect and trigger a further narrowing of the scope of attention. Interestingly, this may be one reason behind why ruminative individuals have faced particular mental health challenges during the COVID-19 pandemic 29 , 30 . Understanding the conscious experience of individuals during times in which they have nothing but their thoughts for distraction may thus have important clinical relevance in differentiating between those for whom internal life is enriching from those for whom it is destructive.

The present study

Quantifying the stream of consciousness for naturally occurring, unprompted thoughts requires a paradigm that captures as much as possible the entirety of thought content in contexts where thoughts are not constrained by experimental task demands. First used to study the thought process of problem solvers and experts 31 , 32 , the think aloud paradigm (TAP) has been used sporadically over the past 50 years to investigate comprehension tasks 33 , 34 , autobiographical memory and future thought 35 , 36 , 37 , 38 , confabulation 39 , and more recently, unprompted thoughts 40 , 41 , 42 . Here we extend beyond prior studies by adapting the oral TAP to explore the basic dynamic properties and ruminative correlates of unprompted thoughts at rest in healthy adults.

Across two studies, we audio recorded participants as they spoke aloud the contents of their conscious experience for 10 min while alone in an unstimulating experimental testing room. Study 1 aimed to assess the validity of the think aloud resting state paradigm and to quantify the basic content and dynamic characteristics of thought in young adults as quantitative benchmarks for Study 2. Motivated by the Broaden and Build theory 22 , 23 , we predicted that thoughts rated as more negative would lead to a narrowing of conceptual scope over time. Study 2 aimed to test the replicability of results from Study 1 and to explore relationships with individual differences in brooding, a maladaptive subtype of rumination 7 , 9 . We hypothesized that higher brooding scores would be associated with more internal, negative, past-oriented, and self-focused thoughts, as well as restricted dynamics and a narrower conceptual scope. Based on the inconsistencies in the literature reviewed above, we were conflicted as to whether thoughts would demonstrate restricted dynamics for negatively-valenced thoughts alone, or whether restricted dynamics with greater trait rumination might also extend to positive thoughts.

Study 1 methods

Participants.

Thirty-three University of Arizona undergraduates participated in Study 1 for monetary compensation (n = 12) or research credit (n = 21). Written informed consent was obtained from all participants and all procedures were performed in accordance with the relevant guidelines and regulations and approved by the University of Arizona’s Institutional Review Board. Five participants had difficulty understanding the task and did not complete the task as instructed. A computer malfunction prevented data collection for one additional participant. The total analyzed sample thus included 27 participants (13 female, mean age = 19.12 years, range = 18–22 years; see Supplementary Table S1 for demographics).

Think aloud paradigm

We used an adapted version of the TAP to audio record participants’ continuous stream of consciousness as individuals sat alone in a testing room for 10 min (Fig.  1 ). Similar to Samson et al. 40 , Sripada and Taxali 41 , and Van Calster et al. 42 , participants were not prompted what to think about. Unlike these studies, participants sat alone in a normally lighted testing room. Similar to these studies, there was minimal external stimulation and the sole instruction was to continuously voice aloud whatever came to their mind, including internal thoughts or images, any perceptions of external stimuli, or any bodily sensations or feelings such as aches, pain, or hunger (similar to a written version of the task in Pennebaker and King 43 ). See Supplementals for complete instructions.

Think aloud paradigm. Participants were audio recorded while voicing aloud their unprompted thoughts for 10 min. Audio recordings were transcribed and coded by hand or automated text analysis for content and dynamics. These indices were explored as predictors of individual differences in trait brooding. MNWT: Mean number of words per thought.

Following the task, participants answered using a continuous sliding scale “How similar were your thoughts to those you experience in your day-to-day life?” and “To what extent did you censor yourself during the task?”. The cursor’s default position was the midpoint between “0-Not at all” and “1-Extremely”. The final location of the cursor was turned into a two-decimal number between 0.00 and 1.00. Participants answered additional questions and completed additional tasks which will be the focus of future manuscripts.

Quantifying individual thoughts, transitions, and themes

Audio files were first transcribed, and three trained independent raters then partitioned the transcript into individual thoughts separated by strong or associative transitions . Thoughts separated by a strong transition demonstrated no obvious content relationship with each other, and the transition felt abrupt (e.g., “ Um, hopefully I’ll get the new job ‘cause I really want it. I’m tired of my old job. [strong transition] Um, I miss my dog. I haven’t seen her in a long time .”). Associational transitions occurred when two thoughts were related in some aspect of content, but nonetheless pertained to different overarching themes, epitomized by the expression “this reminds me of …” (e.g., “Yeah I had Japanese subtitles. I preferred Japanese rather than English. Just sounds very whitewashed when it’s in English [associational transition] But speaking of Japanese, I’m glad I was able to choose what I wanted to do regarding my other language.”) .

This coding scheme provided us with: (a) total word count (overall verbal fluency or the rate with which thoughts are formed and vocalized), (b) total number of thoughts generated (fluency at the level of individual thoughts), (c) total number of strong transitions , (d) total number of associational transitions , and (e) mean number of words averaged across all thoughts (MNW all thoughts , a dynamic estimate of thought duration). Mean number of words per thought was also estimated separately for positive, neutral, and negative thoughts (see below). Raters’ scores were averaged to obtain a mean rating for each variable. Inter-rater reliabilities for each measure ranged from moderate to excellent, with the exception of associational transitions for which it was poor (see Supplementary Table S2 ).

Assessing thought content via manual ratings

After individual thoughts were delineated, each thought was manually rated by two trained independent raters on several content variables. Perceptual orientation was assessed by characterizing each thought as predominantly externally-focused, internally-focused, or interoceptive in nature. Each thought was also manually assigned a predominant temporal orientation (past, present, future, or atemporal), and an overall valence from -5 (extremely negative) to 5 (extremely positive), with 0 = neutral. Degree of self-focus was rated manually for each thought using a scale from 0 (thinking about something entirely unrelated to one’s self) to 4 (clearly and objectively focusing on oneself).

Assessing conceptual scope with semantic similarity

One aspect of dynamics may be captured by how thoughts shift in conceptual scope over time. We thus used natural language processing techniques to estimate the scope of thought using semantic similarity metrics. More similar thoughts likely reflect a narrower conceptual scope over time while less similar thoughts likely reflect a subsequent broadening of thoughts. Using the transcripts manually coded for valence, we computed two metrics of semantic similarity using the spaCy library in Python ( https://spacy.io/ ), one of the fastest and most accurate natural language processing tools available 44 , 45 . Adjacent similarity captures the “movement” between thoughts, or how similar a given thought (thought t ) is to the next thought (thought t  +  1 ). Given that people may return to general topics outside of immediate transitions, we also computed average subsequent similarity , the average similarity between thought t and each subsequent thought (i.e., average [similarity t 1 − t 2 , similarity t 1 − t 3 , … similarity t 1 − t final thought ]). We computed semantic similarity metrics (ranging from 0–1) using a large vector model from spaCy (685 k unique vectors) that derives a “distance” between two thoughts using their mathematical likeness (e.g. “truck” is semantically similar to “car”), after removing unmeaningful “stop” words. Such likeness is determined based on previously established relationships built on text from the web (blogs, news, comments, etc.) and is recommended when a large vocabulary is relevant, as is likely the case for thought data.

Study 1 results

The think aloud paradigm provides an ecologically-valid measure of everyday thinking styles.

Participants reported not to censor themselves very much ( M  = 0.28, SD  = 0.20) and experienced thoughts that were fairly similar to those in their daily lives ( M  = 0.69, SD  = 0.23). On average across participants, the 10-min task yielded 1,216 words ( SD  = 395), in line with what we should expect if participants completed the task as instructed. Considering that speech production for an English speaker is approximately 6 syllables per second 46 and a typical spoken word has a mean of 1.35 syllables 47 , 48 , we can expect 10 min of continuous speech without pauses to yield 2,650 words. However, most humans insert pauses between distinct thoughts and sentences. Participants generated a mean of 28 thoughts ( SD  = 15), comparable to that reported in Sripada and Taxali 41 , with a mean length of 60 words,. Strong transitions represented about two-thirds ( M  = 0.66, SD  = 0.21) of all transitions for each participant. Results are reported in Table 1 . In sum, these findings suggest that the TAP has strong ecological validity.

Content benchmarks are similar to prior studies of everyday thought

As shown in Table 2 , manual ratings of perceptual orientation revealed that 73% ( SD  = 17%) of participants’ thoughts were internal/imaginative, 24% ( SD  = 15%) were external/perceptually-coupled and 3% ( SD  = 3%) pertained to interoceptive experiences such as drowsiness, itch, and hunger, similar to Van Calster et al. 42 . Thoughts were most often rated as present-oriented ( M  = 50%, SD  = 21%), followed by future-oriented ( M  = 21%, SD  = 18%), past-oriented ( M  = 17%, SD  = 11%), and atemporal ( M  = 12%, SD  = 10%). However, the percentage of future-oriented thoughts increased in frequency when internal thoughts were analyzed alone (present: M  = 31%, SD  = 27%; future: M  = 30%, SD  = 26%; past: M  = 23%, SD  = 22%; atemporal: M  = 16%, SD  = 16%). On average, and in line with prior studies 49 , thoughts were rated as predominantly neutral in valence (on scale of − 5 to 5, M  = − 0.15, SD  =  0.4 7). A one sample t-test determined this rating was not statistically different from 0 ( t (26) = − 1.67, P  = 0.11, CI 95 : [− 0.34; 0.04], Cohen’s d  = 0.32). Thoughts were also rated as moderately self-focused ( M  = 2.18, SD  = 0.72, on a scale of 0 to 4).

In summary, thoughts emerging during the TAP tended to be predominantly internal in orientation, moderately self-focused, neutral in valence, and mostly present and prospective in nature. These findings are generally in agreement with prior literature using retrospective questionnaires to assess phenomenological characteristics of thoughts during resting state paradigms 49 , 50 , 51 , 52 .

Increased negative valence leads to a narrowing of conceptual scope

Separate linear mixed effect models were constructed to assess the relationship between adjacent or average subsequent semantic similarity and manual ratings of valence for each thought. For these models, manually-rated valence was the dependent variable, and semantic similarity was the fixed effect of interest. We also included a random intercept to allow for baseline variability in valence across participants. When including random slopes in our model, they did not converge, suggesting our data does not support this more complex model structure.

The relationship between adjacent similarity and valence was statistically significant ( b  = − 0.59, χ 2 (1) = 4.78, P  = 0.029), revealing a generally negative pattern ( ß  = − 0.08 (CI 95 : [− 0.17; − 0.01]). As thoughts became more negative, they became more semantically similar to the next thought. Conversely, as thoughts became more positive, they became more semantically distant to the next thought. Similar findings were observed with the average subsequent similarity metric ( b  = − 1.12, χ 2 (1) = 9.49, P  = 0.002, ß  = − 0.16 (CI 95 : [− 0.26; − 0.06]), where more negative thoughts were more semantically similar to the average of all subsequent thoughts. These findings suggest negative thoughts are associated with a subsequent narrowing of conceptual scope while positive thoughts are associated with a subsequent broadening of conceptual scope.

Study 1 discussion

Overall, the results of Study 1 make three important contributions to the literature. First, most participants did not appear to experience the TAP as overtly artificial, and were able to experience and vocalize thoughts that were fairly representative of those experienced in daily life. Thought content was also in general agreement with the prior literature; thoughts were moderately self-focused, neutral in valence, and mostly pertained to the present and future. Beyond content, the TAP offers additional indices of dynamics, including the overall number and length of thoughts that can be explored separately for positive and negative thoughts. Third, although most thoughts were characterized by abrupt transitions, the conceptual similarity or scope of thoughts over time is moderated by thought valence. More negative thoughts were associated with a subsequent narrowing of the conceptual scope of attention.

To test the replicability of these results and examine how stream of consciousness indices vary by trait rumination, we conducted a second study using identical procedures as Study 1. Three important additions to Study 2 were: 1) the measurement and analysis of individual differences in trait brooding, 2) content coding of thoughts using automated text-analysis software, and 3) the computation of affective transition probabilities between affective thoughts.

Study 2 methods

Fifty-seven University of Arizona undergraduates with high English proficiency participated in Study 2 in exchange for class credit. Written informed consent was obtained from all participants and all procedures were performed in accordance with the relevant guidelines and regulations and approved by the University of Arizona’s Institutional Review Board. Six participants were excluded as a result of failure to comply with the instructions of the task (n = 3), poor sound quality (n = 1), and a computer malfunction in which data were not collected (n = 2). The final sample consisted of 51 participants (32 females, 1 non-binary, M  = 19.78 years, range = 18–28 years). Demographic details are listed in Supplementary Table S1 .

Materials and procedures

Participants in Study 2 completed an identical TAP as Study 1 in the same room and with identical instructions. The audio files were transcribed in the same way and by the same researchers as in Study 1. To estimate individual differences in trait rumination, participants also completed the Rumination Response Scale (RRS) 9 . As we were most interested in the dynamics of maladaptive rumination 9 , we used scores from the 5-item RRS-brooding subscale for subsequent analyses 53 which has good reliability (α = 0.77) 9 . Participants’ brooding scores ranged from 6 to 16, covering much of the scale (5 to 20) ( M  = 10, SD  = 2.93).

A key aim of Study 2 was to determine the replicability of results from Study 1; therefore, we implemented identical data preparation and analysis procedures. The intra-class correlation coefficients for variables measured across the two studies were similar (see Supplementary Table S2 ) and remained within the same interpretative category (e.g. excellent, good, moderate, poor).

Automatic text analysis

In addition to manual coding, content was also assessed automatically using one of the most widely used and extensively validated dictionary-based text-analysis applications: Linguistic Inquiry Word Count (LIWC) v.2015 54 , 55 . LIWC coding offered ease and rapidity of implementation and allowed us to estimate its potential for scalability in future studies. LIWC calculates the proportion of words in the transcripts belonging to a given number of linguistic categories. Here we focused on the proportion of positive, negative, past-oriented, and future-oriented words used, as well as self-focus assessed with first person pronoun usage (e.g. I, me, my), which has reliably been linked to depression and negative emotionality 56 .

Statistical analysis

Partial correlations between brooding scores and indices spanning content and dynamics were run, controlling for individual differences in perceived censorship and similarity of thoughts to daily life (for first order Pearson correlations, see Supplementary Table S3 ). For models including dynamic indices, analyses additionally controlled for total word count. One data point with high leverage (Cook d > 0.50) was removed for two analyses (manually-coded past and MNW all thoughts ). No analysis violated assumptions for linear modelling analysis.

As in Study 1, content analyses focused on perceptual orientation, valence, self-focus, and temporal orientation. For dynamics, we examined relationships between trait brooding and the total number of thoughts, as well as three duration indices: MNW all thoughts , MNW negative thoughts and MNW positive thoughts (though we also report MNW neutral thoughts for completeness). Each thought was classified as positive when its average valence rating across the two raters was ≥ 1 (on a scale of − 5 to 5), negative when its valence rating was ≤ − 1, and neutral when its rating was [− 0.99 to 0.99]. We tested whether the various benchmarks measured in Study 1 differed in Study 2 with two-tailed Welch t-test or Wilcoxon rank-sum test depending on whether the variable distribution were normally distributed as determined by a Shapiro–Wilk test. The Welch t-test was used for all Study 1 vs. Study 2 comparisons that did not violate the distribution normality assumptions because it controls for Type 1 error better than the Student’s t-test in case of unequal variances between the groups and has been shown to be just as robust when this assumption is met 57 .

We also repeated the analyses from Study 1 to determine whether negative valence was once again related to the conceptual scope of subsequent thoughts. In light of our hypotheses, we also assessed whether brooding scores moderated this relationship.

In addition, we calculated affective transition probabilities; the likelihood that for each thought of a particular valence (Positive, Neutral, Negative), the next subsequent thought will remain in the same valence state (e.g. Pos → Pos, Neut → Neut, Neg → Neg) or shift to one of the other states (e.g. Pos → Neut, Pos → Neg, etc.). Affective transition probabilities were calculated separately across all positive thoughts, all neutral thoughts, and all negative thoughts. This method resulted in a 3 × 3 table with all possible transitional states (see Supplementary Table S4 for descriptive statistics).

Finally, we performed nested model comparisons to determine whether the dynamic indices explained variance in brooding scores above and beyond that explained by content alone. We compared a) a baseline model with all significant content variables (separate models for manual and automated coding), censorship, and daily thought similarity, to b) an intermediate model adding the total word count (to account for the contribution of verbal fluency), to c) a final model including content, censorship, daily thought similarity, total word count, and all significant dynamic variables significantly related to brooding scores (MNW positive , MNW negative , and affective transition probabilities from Pos → Pos and Neg → Pos).

Study 2 results

Study 2 replicated the findings of study 1.

All but one of the indices explored in Study 1 were statistically similar to those obtained in Study 2 (Tables 1 and 2 ). Participants reported censoring themselves during the task more in Study 2 than Study 1 ( t (67.62) = − 2.43, P  = 0.012, Cohen’s d  = 0.53). Also, there was a noticeable though non-significant increase in future orientation in Study 2 ( P  = 0.076). Restricting the temporal analysis to internally generated thoughts, future orientation became the most prominent temporal orientation in Study 2 (39% future, 23% present, 17% past, 12% atemporal). In summary, results from Study 2 largely mimic those from Study 1, suggesting the task is ecologically-valid and reliable across two independent samples.

Trait brooding was associated with more negative, past oriented, and self-focused thinking

As hypothesized, high brooding scores were associated with thoughts that were more negative and past-oriented for both manual ratings and automated text analysis (all p ’s < 0.04, Table 3 ). In line with our predictions, trait brooding was also linked to a higher frequency of first-person pronoun usage. However, manual ratings of self-focus did not relate to trait brooding ( partial correlation  = 0.19, P  = 0.19).

High brooding scores were associated with longer negative and shorter positive thoughts

Collapsing across affective content, dynamic indices did not predict individual differences in trait brooding. Trait brooding was unrelated to the total number of thoughts, the number of strong transitions, the number of associative transitions, or the mean number of words of thoughts (all p ’s > 0.36). However, as trait brooding scores increased, participants produced fewer words overall across the 10-min TAP ( partial correlation  = − 0.29, P  = 0.041).

Nevertheless, relationships with trait brooding became apparent when separating thoughts by their affective content. MNW negative was significantly positively associated with brooding scores ( partial correlation  = 0.36, P  = 0.010), and MNW positive was numerically but non-significantly negatively associated with brooding ( partial correlation  = − 0.27, P  = 0.051). MNW neutral bore no relationship with brooding scores ( partial correlation  = − 0.08, P  = 0.56).

Since the duration of positive and negative thoughts was not strongly correlated ( r  = 0.11, P  = 0.44), we included both measures in the same model predicting trait brooding. The model explained 28.4% ( adjusted r 2 ) of the variance in brooding scores ( F (5,45) = 4.96, P  = 0.001), and both dynamic variables uniquely contributing (partial correlations similar to those reported Table 3 ). Overall, as brooding scores increased, positive thoughts became shorter and negative thoughts became longer.

Trait brooding tends to repel positive conceptual states: Affective transition probability analyses

Higher brooding was associated with a lower likelihood of remaining in subsequent positive conceptual states (Pos → Pos: partial r − 0.31, P  = 0.025) and a lower likelihood of transitioning from a negative thought to a positive thought (Neg → Pos: partial r  = − 0.32, P  = 0.021) (Fig.  2 ). Contrary to expectations, trait brooding was not associated with a stronger tendency to remain in a negative conceptual state following a negative thought (Neg → Neg: partial r  = 0.09, P  = 0.53). All other transition types were unrelated to brooding scores (all P ’s > 0.21).

Linear relationships between affective transition probabilities and trait brooding. The probabilities of transitioning to a positive (green), neutral (grey), or negative (red) thought from current thoughts that are positive (top), neutral (middle), or negative (bottom) are examined in relationship to trait brooding. The y-axis reflects brooding once controlling for the covariates of censorship and daily thought similarity. * P  < .05.

Brooding moderates the relationship between negative thought and average subsequent semantic similarity

In contrast to Study 1, the relationship between adjacent similarity and valence was not statistically significant, b  = − 0.092, χ 2 (1) = 0.166, P  = 0.68, ß  = − 0.01 ( CI 95 : [− 0.06; 0.04]). However, replicating the results from Study 1, average subsequent similarity was significantly negatively related to valence, b  = − 0.766, χ 2 (1) = 5.87, P  = 0.015, ß  = − 0.07 ( CI 95 : [− 0.13; − 0.01]).

We then constructed two regression models to assess if the relationship between thought similarity and valence was moderated by brooding scores. Similarity (adjacent or average subsequent, respectively), brooding score, and their interaction term were included as fixed effects, with a random effect (intercept) of participant. Both models controlled for individual differences in perceived thought censorship, similarity with daily life, and total word count.

We observed a significant interaction between brooding and subsequent similarity (Fig.  3 ), b  = − 0.26, χ 2 (1) = 5.89, P  = 0.015, suggesting that brooding scores moderated the relationship between valence and similarity over time. For those higher in brooding, greater negative valence was associated with more conceptual similarity over time, consistent with a progressive narrowing of conceptual scope (Fig.  3 ). Simple slopes analyses revealed that subsequent similarity was significantly related to valence when brooding scores were high (+ 1 SD above the mean), b  = − 1.05, P  = 0.030, whereas the same relationship was not significant for participants at the mean or − 1 SD below the mean ( P ’s = 0.32 and 0.40, respectively). The interaction was not significant for adjacent ( b  = − 0.137, χ 2 (1) = 3.05, P  = 0.08), suggesting that this relationship is stronger when considering how thoughts unfold over time and not simply accounting for a single thought transition. Finally, we repeated both models without the covariates (similarity with daily life, censorship, and similarity total word count) and the pattern of significant results was consistent.

Brooding scores moderate the relationship between thought valence and average semantic similarity across subsequent thoughts. More negatively-valenced thoughts led to greater average subsequent semantic similarity (narrower conceptual scope) for individuals with higher brooding tendencies . The simple slopes relationships between valence and average subsequent semantic similarity for low and mean brooding groups was not significant. * P  < .05.

Thought content and dynamics account for unique variance in predicting trait brooding

In our nested model comparisons (Table 4 ), valence and % past-orientation were included as content predictors, and MNW positive , MNW negative , and affective transition probabilities Pos → Pos and Neg → Pos were included as dynamic indices. All models were significant (all p ’s < 0.02). Content, censorship and similarity to daily life (Model 1) collectively explained 17.73% of the variance in trait brooding, and the addition of total word count (Model 2) explained significantly more variance than the initial model ( χ 2 (1) = 7.21, P  = 0.010, Δadjusted r 2  = 8.35%, ΔAIC  = 4.6). Adding dynamic predictors in Model 3 explained 10% more variance in brooding than Model 2 ( χ 2 (4) = 2.83, P  = 0.037, Δadjusted r 2  = 10.36%, ΔAIC  = 4.4), explaining a total of 36.44% of variance in brooding from a ten-minute TAP.

A similar analysis was performed using the LIWC-derived significant predictors (i.e., % of negative, past-oriented, and first-person pronoun words). Again, all models were significant (all P ’s < .0.003). Though Model 2 explained significantly more variance than Model 1 ( χ 2 (1) = 6.29, P  = 0.016, Δadjusted r 2  = 7.17%, ΔAIC  = 4.2), the increase in variance explained by the Model 3 was not significantly greater than the intermediate model ( χ 2 (4) = 2.20, P  = 0.086, Δadjusted r 2  = 6.78%, ΔAIC  = 2.1).

Study 2 discussion

Metrics derived from manual and LIWC coding of participants’ thoughts were remarkably similar across both studies. Thoughts sampled during an extended period of rest were rated as ecologically valid and showed similarity in content to studies reported in the literature using retrospective and experience sampling methods. On average, thoughts were neutral in valence 49 , 58 , and internal thoughts had a prospective bias 52 . Consistent with prior work suggesting that negative thinking is associated with more inflexibility 16 , 19 , 22 , 23 , 24 , 59 , negative thoughts further narrowed the conceptual scope of subsequent thoughts. This successful cross-validation provides a strong argument for the reliability of the task.

Applying the TAP to investigate brooding tendencies in young adults, higher brooding scores were associated with more negative, past oriented, and self-referential language. Those results mimic those found in clinically and subclinically depressed individuals 60 , 61 . As brooding increased, negative thoughts became longer and positive thoughts became shorter, and brooding tended to repel participants away from positive thoughts. Content and dynamic indices explained independent variance toward predicting brooding, and together accounted for more than a third of the adjusted r 2 variability in trait brooding. These different results support the ecological, convergent, and clinical validity of the TAP.

General discussion

Across two studies in independent samples of young adults, we audio recorded streams of consciousness during a 10-minute resting state paradigm and quantified metrics of content, dynamics, conceptual scope, and overall task experience. Study 1 provided quantitative benchmarks for these metrics and revealed the paradigm’s strong ecological validity. Study 2 replicated these metrics and related them to brooding scores, offering a cognitive signature of trait brooding during idle time. Collectively, these findings have important clinical and basic science implications and open up a range of exciting possibilities for future research.

High brooding tendencies associate with altered thought content and dynamics at rest

Although dysfunctional content, dynamics, and conceptual scope have been recognized as central to many mental health disorders 4 , 5 , 6 , 10 , 24 , metrics beyond content have rarely been quantitatively explored, much less together in a single study. Similar to our study, Sripada and Taxali 41 asked participants to “think aloud” for 30 min at rest and subsequently isolated individual thoughts. Although the authors did not extend their findings to individual differences in mental health, the stream of consciousness in healthy individuals tended to “clump” together by semantically similar content and then “jump” to different topics. Though our coding scheme differed, we similarly observed a clump-and-jump tendency, as suggested by the predominance of strong transitions (~ 73%). Additionally, Molina and colleagues 51 used a TAP to examine changes in coarse linguistic content and topic shifting from a baseline period to a worry-prompted state across anxious, depressed, and healthy adults. Compared to the unprompted TAP, the worry TAP was characterized by more negative content and less shifts from topic to topic, although no group differences were observed. Relatedly, following a ruminative prompt, Lyubomirsky and colleagues 62 found that dysphoric participants have a stronger tendency to reflect repeatedly on their problems. Here we extend beyond these prior studies by showing that in the absence of an overt mood induction or prompt, individuals with ruminative tendencies generate continuous thoughts that differ in both content and dynamics.

Our observation of restricted thought dynamics in individuals prone to brooding is consistent with a recent theoretical proposal that rumination is a form of thought constrained by a strong automatic affective pull 10 , 11 . Instead of freely moving from topic to topic, individuals prone to rumination exhibited longer negative thoughts, and negative content in ruminative individuals was linked to a stronger narrowing in conceptual scope over time, as suggested by higher levels of semantic similarity. Importantly, these dynamic indices explained variability in trait brooding beyond content and overall task experience. As such, they also converge with a recent study using a chained free association task showing that the conceptual associations of individuals with higher trait rumination were more strongly attracted to negative conceptual spaces and more likely to remain there longer 59 . Collectively, these studies provide behavioral support for the role of automatic constraints in restricting thought flexibility 10 , 14 , and are worthy of further empirical exploration given their predictive validity.

Although the overall number of thoughts and transitions did not differ across brooding scores, higher trait brooding was associated with restricted dynamics for negative compared to positive thoughts. These findings are broadly consistent with attentional scope models such as the Broaden-and-Build theory 22 , 23 and aspects of the Attentional Scope Model of Rumination 24 (see also 63 ) and relate to prior findings showing that negative mood is associated with decreases in divergent thinking 64 . Negative thoughts were associated with a further narrowing of conceptual scope, and this relationship became stronger with increased trait brooding. Unlike extended predictions of the Attentional Scope Model of Rumination 24 , however, this narrowing of conceptual scope did not extend to positive thoughts in ruminative individuals. Additionally, the affective transition probability analysis did not suggest a propensity to stick to a given thought valence, with positive-to-positive transitions being significantly less likely as brooding scores increased. These findings contrast with the emotional inertia literature 20 , although emotional inertia pertains to the evolution of state affect 19 , while our results consider the evolution of affective content of thought and semantic similarity of content over time. Following from these findings, a speculative prediction would be that individuals high in trait brooding may perform particularly well on executive function tasks that require maintenance of attention towards negative stimuli, or switching from positively-valenced to negatively-valenced tasks. However, when such stimuli is no longer relevant for the task at hand, individuals high in brooding may exhibit difficulties disengaging from such stimuli or unrelated ruminative thought impairing task performance 65 .

We also observed that higher trait brooding associated with fewer words across the 10-min resting state paradigm. Although total word count was accounted for in our models of thought dynamics, the underlying cause of this lower word count in rumination remains unclear. We considered it may be related to higher censorship. Trait brooding was indeed marginally linked to higher thought censorship ( P  = 0.052), yet word count and censorship were not significantly correlated across participants (Study 1: r  = − 0.20, P  = 0.17; Study 2: r  = − 0.13, P  = 0.50), suggesting that higher brooding may be independently linked to both reduced overall thought fluency and somewhat greater censorship. Reduced word count in individuals with higher brooding could thus reflect difficulties in translating thoughts into spoken words (akin to verbal fluency), or an overall slowing of the rate of thought (i.e., slow ideation), both of which have been related to depression 66 , 67 , 68 .

Thoughts emerging at rest are a clinically important yet empirically neglected aspect of mental life

Although the TAP can be used to capture thoughts during a task or following a prompt, here we capture thoughts emerging unprompted and at “rest,” when individuals are left alone with their thoughts with few opportunities for distraction. This aspect of our mental life requires more investigative effort for its potential relevance to mental health, as these moments may be most auspicious for the maladaptive aspects of thought 26 , 27 , 28 . Indeed, brooding being a state of high self-absorption associated with high level of decoupling from the external environment, it is conceivable that moments providing minimal opportunity for external engagement and distraction may act as contextual and mnemonic cues triggering habitual perceptually-decoupled brooding thought. It is also possible that negatively-valenced or self-related external stimuli can trigger the initiation of a chain of habitual ruminative thinking through associative processes 59 , 69 . How ruminative trains of thought begin is an important avenue for future investigation. Of relevance are mindfulness practices which train practitioners to face whatever thought content emerges with equanimity and to be comfortable being alone with their mind 70 , 71 . Additionally, the practice trains attentional focus 72 by strengthening attentional control networks 73 , such that experienced meditators are more resilient to distraction and intrusive thoughts. Accordingly, meditation experience has been linked to greater deactivation of the default mode network during a variety of meditation practices 74 , and mindfulness-based therapies have been shown to reduce symptoms of depression and anxiety 75 , 76 .

These findings also provide insight into the cognitive mechanisms underpinning the clinical predictive validity of resting state functional MRI connectivity paradigms 77 , 78 . While the majority of resting state studies do not assess the content of thought emerging at rest, our findings suggest that the content and dynamics of thought explain a large amount of variability in ruminative traits between individuals. A prediction of our work is that this variability may also contribute to differences in resting state connectivity patterns between individuals, as well as variability within individuals, as suggested by recent studies 42 , 79 , 80 , 81 , 82 , 83 , 84 . For example, Turnbull and colleagues showed that as individuals perform an external task, task positive brain networks progressively decrease their activity with the passage of time, and individual differences in these neural dynamics track individual differences in self-reported task-unrelated thought 83 . Additionally, variation in neural dynamics across individuals have been linked to individual differences in fluid intelligence and divergent thinking, as well as one’s ability to regulate the emergence of task-unrelated thoughts based on task demands 84 . Overall, we suggest the think aloud paradigm may complement and extend this more recent research by affording a more precise mapping of thought content to the dynamics of hemodynamic activity.

Advantages of the think aloud paradigm to study the stream of consciousness

As demonstrated here and by others 85 , 86 , the TAP is a short, convenient, and ecologically-valid paradigm with convergent and predictive validity that significantly adds to our understanding of thought content and dynamics. Additionally, it has the advantage of circumventing issues related to retrospective and subjective assessment of thought, which is common in resting state paradigms. Our findings using automated LIWC text analysis software also highlight the potential scalability of the TAP. Though manual ratings are considered more trustworthy, they are time intensive and require multiple raters. Automated and manual methods of assessment yielded similar content across Study 1 and 2 ( r   ≅  0.50), and the Pearson and partial correlations with brooding scores were also quite similar, except for the manually coded level of self-focus of thought. The latter may be explained by the possibility that self-references and self-focus may tap into slightly different constructs. Although the time-consuming nature of our manual coding procedure limited our sample sizes in Study 1 and 2, future research would be benefitted by applying linguistic software such as LIWC to examine think aloud transcripts in much larger samples.

Study limitations and future directions

This study has some limitations. Aside from being a predominantly young, educated, and relatively small sample, the laboratory environment in which the study took place may have felt unusual to the participants and prompted some levels of inhibition, affecting the quantity and content of thought produced. Additionally, we did not use a social desirability scale to determine whether participants’ answers to the ecological validity questions (i.e., censorship and similarity to daily life) were sincere. However, multiple findings including thought fluency analyses, as well as the presence of content including self-disclosure, self- or other-focused criticism, cursing and so on converge to suggest that participants’ transcripts were largely sincere.

Another limitation is that we did not assess state affect during the experiment and thus could not examine the effect of mood on our findings. Additionally, our participants were skewed towards exhibiting lower levels of brooding, although the range in scores spanned nearly the full scale. A final limitation is that many of our statistical effects are relatively modest and would likely not survive corrections for multiple comparisons, especially considering our relatively small sample sizes. However, findings pertaining to variables other than rumination replicated across two studies, were consistent across measurement techniques (manual and LIWC), and were predicted a priori (with the exception of the duration analyses for which we also acknowledged an alternative hypothesis). Nonetheless, we call for future research to replicate our findings in larger sample sizes, including in clinically-diagnosed samples.

An important avenue for future work will therefore be to extend the TAP to clinical samples with higher levels of rumination. From the intervention side, comparing the content of thoughts prior, during, and after clinical interventions may also be an avenue of interest 87 , with potential to illuminate the evolution of the stream of consciousness as symptoms change over time. Despite these limitations, the TAP is a promising paradigm with basic science, clinical and therapeutic relevance.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

James, W. The Principles of Psychology (Henry Holt & Co., 1890).

Google Scholar  

Andrews-Hanna, J. R. & Grilli, M. D. Mapping the imaginative mind: charting new paths forward. Curr. Dir. Psychol. Sci. 30 , 82–89 (2021).

Article   PubMed   PubMed Central   Google Scholar  

Baars, B. J. Spontaneous repetitive thoughts can be adaptive: postscript on ‘mind wandering’. Psychol. Bull. 136 , 208–210 (2010).

Article   PubMed   Google Scholar  

Watkins, E. R. Constructive and unconstructive repetitive thought. Psychol. Bull. 134 , 163–206 (2008).

Ehring, T. & Watkins, E. R. Repetitive negative thinking as a transdiagnostic process. Int. J. Cogn. Ther. 1 , 192–205 (2008).

Article   Google Scholar  

Kaplan, D. M. et al. Maladaptive repetitive thought as a transdiagnostic phenomenon and treatment target: an integrative review. J. Clin. Psychol. 74 , 1126–1136 (2018).

Aldao, A., Nolen-Hoeksema, S. & Schweizer, S. Emotion-regulation strategies across psychopathology: a meta-analytic review. Clin. Psychol. Rev. 30 , 217–237 (2010).

Nolen-Hoeksema, S., Wisco, B. E. & Lyubomirsky, S. Rethinking rumination. Perspect. Psychol. Sci. 3 , 400–424 (2008).

Treynor, W., Gonzalez, R. & Nolen-Hoeksema, S. Rumination reconsidered: a psychometric analysis. Cognit. Ther. Res. 27 , 247–259 (2003).

Andrews-Hanna, J. R., Christoff, K. & O’Connor, M.-F. Dynamic regulation of internal experience: mechanisms of therapeutic change. In The Neuroscience of Enduring Change: The Neural Basis of Talk Therapies (eds Lane, R. et al. ) 89–191 (Oxford University Press, 2020).

Chapter   Google Scholar  

Christoff, K., Irving, Z. C., Fox, K. C. R., Spreng, R. N. & Andrews-Hanna, J. R. Mind-wandering as spontaneous thought: a dynamic framework. Nat. Rev. Neurosci. 17 , 718–731 (2016).

Article   CAS   PubMed   Google Scholar  

Irving, Z. C. Mind-wandering is unguided attention: accounting for the “purposeful” wanderer. Philos. Stud. 173 , 547–571 (2016).

DuPre, E. & Spreng, R. N. Rumination is a sticky form of spontaneous thought. In The Oxford Handbook of Spontaneous Thought (eds Fox, K. C. R. & Christoff, K.) 509–520 (Oxford University Press, 2018).

Girn, M., Mills, C., Roseman, L., Carhart-Harris, R. L. & Christoff, K. Updating the dynamic framework of thought: creativity and psychedelics. Neuroimage 213 , 116726 (2020).

Mills, C., Raffaelli, Q., Irving, Z. C., Stan, D. & Christoff, K. Is an off-task mind a freely-moving mind? Examining the relationship between different dimensions of thought. Conscious. Cogn. 58 , 20–33 (2018).

Ottaviani, C., Shapiro, D. & Couyoumdjian, A. Flexibility as the key for somatic health: from mind wandering to perseverative cognition. Biol. Psychol. 94 , 38–43 (2013).

Moberly, N. J. & Watkins, E. R. Ruminative self-focus and negative affect: an experience sampling study. J. Abnorm. Psychol. 117 , 314–323 (2008).

Mills, C., Porter, A., Andrews-Hanna, J., Christoff, K. & Colby, A. How task-unrelated and freely-moving thought relate to affect: evidence for dissociable patterns in everyday life. Cogn. Emot. 6 , 17 (2020).

Kuppens, P., Allen, N. B. & Sheeber, L. Emotional inertia and psychological maladjustment. Psychol. Sci. 21 , 984–991 (2010).

Koval, P., Sütterlin, S. & Kuppens, P. Emotional inertia is associated with lower well-being when controlling for differences in emotional context. Front. Psychol. 6 , 1–11 (2016).

Houben, M., Van Den Noortgate, W. & Kuppens, P. The relation between short term emotion dynamics and psychological well-being: a meta-analysis. Psychol. Bull. 141 , 901–930 (2015).

Fredrickson, B. L. The broaden-and-build theory of positive emotions. Philos. Trans. R. Soc. B Biol. Sci. 359 , 1367–1377 (2004).

Garland, E. L. et al. Upward spirals of positive emotions counter downward spirals of negativity: insights from the broaden-and-build theory and affective neuroscience on the treatment of emotion dysfunctions and deficits in psychopathology. Clin. Psychol. Rev. 30 , 849–864 (2010).

Whitmer, A. J. & Gotlib, I. H. An attentional scope model of rumination. Psychol. Bull. 139 , 1036–1061 (2013).

Altamirano, L. J., Miyake, A. & Whitmer, A. J. When mental inflexibility facilitates executive control: beneficial side effects of ruminative tendencies on goal maintenance. Psychol. Sci. 21 , 1377–1382 (2010).

Solomon, P. Quantitative aspects in sensory deprivation. In The Psychodynamic Implications of Physiological Studies on Sensory Deprivation (eds Snow, L. M. & Snow, L. H.) (Thomas, 1970).

Vernon, J. A. Inside the Black Box . (Clarkson Potter, 1963).

Wilson, T. D. et al. Just think: the challenges of the disengaged mind. Science 345 , 75–77 (2014).

Article   ADS   CAS   PubMed   PubMed Central   Google Scholar  

Hong, W. et al. Social media exposure and college students’ mental health during the outbreak of COVID-19: the mediating role of rumination and the moderating role of mindfulness. Cyberpsychol. Behav. Soc. Netw. 24 , 282–287 (2020).

Satici, B., Saricali, M., Satici, S. A. & Griffiths, M. D. Intolerance of uncertainty and mental wellbeing: serial mediation by rumination and fear of COVID-19. Int. J. Ment. Health Addict. 6 , 1. https://doi.org/10.1007/s11469-020-00305-0 (2020).

De Groot, A. D. Thought and Choice in Chess (Revised translation of De Groot, 1946) . (Ishi Press, 2016). https://doi.org/10.1515/9783110816877-008 .

Duncker, K. On Problem Solving. (The American Psychological Association, Washington., 1945).

Bohn-Gettler, C. M. & Kendeou, P. The interplay of reader goals, working memory, and text structure during reading. Contemp. Educ. Psychol. 39 , 206–219 (2014).

Magliano, J. P. & Millis, K. K. Assessing reading skill with a think-aloud procedure and latent semantic analysis. Cogn. Instr. 21 , 251–283 (2003).

D’Argembeau, A. & Mathy, A. Tracking the construction of episodic future thoughts. J. Exp. Psychol. Gen. 140 , 258–271 (2011).

Haque, S. & Conway, M. A. Sampling the process of autobiographical memory construction. Eur. J. Cogn. Psychol. 13 , 529–547 (2001).

Gilmore, A. W. et al. Evidence supporting a time-limited hippocampal role in retrieving autobiographical memories. Proc. Natl. Acad. Sci. U. S. A. 118 , 11 (2021).

Article   CAS   Google Scholar  

Ben, M. H. et al. Temporal processing of past and future autobiographical events in patients with schizophrenia. Sci. Rep. 9 , 1–11 (2019).

ADS   Google Scholar  

Burgess, P. W. & Shallice, T. Confabulation and the control of recollection. Memory 4 , 359–412 (1996).

Samson, A., Simpson, D., Kamphoff, C. & Langlier, A. Think aloud: an examination of distance runners’ thought processes. Int. J. Sport Exerc. Psychol. 15 , 176–189 (2017).

Sripada, C. & Taxali, A. Structure in the stream of consciousness: evidence from a verbalized thought protocol and automated text analytic methods. Conscious. Cogn. 85 , 16 (2020).

Van Calster, L., D’Argembeau, A., Salmon, E., Peters, F. & Majerus, S. Fluctuations of attentional networks and default mode network during the resting state reflect variations in cognitive states: evidence from a novel resting-state experience sampling method. J. Cogn. Neurosci. 29 , 95–113 (2017).

Pennebaker, J. W. & King, L. A. Linguistic styles: language use as an individual difference. J. Pers. Soc. Psychol. 77 , 1296–1312 (1999).

Al Omran, F. N. A. & Treude, C. Choosing an NLP library for analyzing software documentation: a systematic literature review and a series of experiments. in IEEE International Working Conference on Mining Software Repositories 187–197 (IEEE Computer Society, 2017). doi: https://doi.org/10.1109/MSR.2017.42 .

Choi, J. D., Tetreault, J. & Stent, A. It depends: dependency parser comparison using a web-based evaluation tool. ACL-IJCNLP 2015-53rd Annu. Meet. Assoc. Comput. Linguist. 7th Int. Jt. Conf. Nat. Lang. Process. Asian Fed. Nat. Lang. Process. Proc. Conf. 1 , 387–396 (2015).

Coupé, C., Oh, Y., Dediu, D. & Pellegrino, F. Different languages, similar encoding efficiency: comparable information rates across the human communicative niche. Sci. Adv. 5 , 19 (2019).

Andrews, G. & Ingham, R. J. Stuttering: consideration in the evaluation of treatment. Int. J. Lang. Commun. Disord. 6 , 129–138 (1971).

Venkatagiri, H. S. Clinical measurement of rate of reading and discourse in young adults. J. Fluency Disord. 24 , 209–226 (1999).

Fox, K. C. R. et al. Affective neuroscience of self-generated thought. Ann. N.Y. Acad. Sci 1426 , 25–51 (2018).

Article   ADS   Google Scholar  

Diaz, A. B. et al. The Amsterdam Resting-state Questionnaire reveals multiple phenotypes of resting-state cognition. Front. Hum. Neurosci. 7 , 1–15 (2013).

Molina, S., Borkovec, T. D., Peasley, C. & Person, D. Content analysis of worrisome streams of consciousness in anxious and dysphoric participants. Cognit. Ther. Res. 22 , 109–123 (1998).

Smallwood, J. & Schooler, J. W. The science of mind wandering: empirically navigating the stream of consciousness. Annu. Rev. Psychol. 66 , 487–518 (2015).

Watkins, E. R. Depressive rumination and co-morbidity: evidence for brooding as a transdiagnostic process. J. Ration. Emotive Cogn. Behav. Ther. 27 , 160–175 (2009).

Boyd, R. L. & Schwartz, H. A. Natural language analysis and the psychology of verbal behavior: the past, present, and future states of the field. J. Lang. Soc. Psychol. 40 , 21–41 (2021).

Tausczik, Y. R. & Pennebaker, J. W. The psychological meaning of words: LIWC and computerized text analysis methods. J. Lang. Soc. Psychol. 29 , 24–54 (2010).

Tackman, A. M. et al. Depression, negative emotionality, and self-referential language: a multi-lab, multi-measure, and multi-language-task research synthesis. J. Pers. Soc. Psychol. 116 , 817–834 (2019).

Delacre, M., Lakens, D. & Leys, C. Why psychologists should by default use welch’s t-Test instead of student’s t-Test. Int. Rev. Soc. Psychol. 30 , 92–101 (2017).

Trampe, D., Quoidbach, J. & Taquet, M. Emotions in everyday life. PLoS ONE 10 , 145450 (2015).

Andrews-Hanna, J. R. et al. The conceptual building blocks of everyday thought: tracking the emergence and dynamics of ruminative and non-ruminative thinking. J. Exp. Psychol. Gen.

Hoffmann, F., Banzhaf, C., Kanske, P., Bermpohl, F. & Singer, T. Where the depressed mind wanders: self-generated thought patterns as assessed through experience sampling as a state marker of depression. J. Affect. Disord. 198 , 127–134 (2016).

Andrews-Hanna, J. R. et al. A penny for your thoughts: dimensions of self-generated thought content and relationships with individual differences in emotional wellbeing. Front. Psychol. 4 , 1–13 (2013).

Lyubomirsky, S., Tucker, K. L., Caldwell, N. D. & Berg, K. Why ruminators are poor problem solvers: clues from the phenomenology of dysphoric rumination. J. Pers. Soc. Psychol. 77 , 1041–1060 (1999).

Grol, M., Hertel, P. T., Koster, E. H. W. & De Raedt, R. The effects of rumination induction on attentional breadth for self-related information. Clin. Psychol. Sci. 3 , 607–618 (2015).

Baas, M., De Dreu, C. K. W. & Nijstad, B. A. A meta-analysis of 25 years of mood-creativity research: hedonic tone, activation, or regulatory focus?. Psychol. Bull. 134 , 779–806 (2008).

Gotlib, I. H. & Joormann, J. Cognition and depression: current status and future directions. Annu. Rev. Clin. Psychol. 6 , 285–312 (2010).

Lee, R. S. C., Hermens, D. F., Porter, M. A. & Redoblado-Hodge, M. A. A meta-analysis of cognitive deficits in first-episode Major Depressive Disorder. J. Affect. Disord. 140 , 113–124 (2012).

Bennabi, D., Vandel, P., Papaxanthis, C., Pozzo, T. & Haffen, E. Psychomotor retardation in depression: a systematic review of diagnostic, pathophysiologic, and therapeutic implications. Biomed Res. Int. 2013 , (2013).

Cummins, N. et al. A review of depression and suicide risk assessment using speech analysis. Speech Commun. 71 , 10–49 (2015).

Bower, G. H. Mood and memory. Am. Psychol. 36 , 129–148 (1981).

Kabat-Zinn, J. Wherever You Go, There you are: Mindfulness Meditation in Everyday Life. (Hyperion, 1994).

Segal, Z. V, Teasdale, D., J. & Williams, J. M. G. Mindfulness-based cognitive therapy: theoretical rationale and empirical status. in Mindfulness and Acceptance: Expanding the Cognitive-Behavioral Tradition (eds. Hayes, S. C., Follette, V. M. & Linehan, M. M.) 45–65 (Guilford Press., 2004).

Sumantry, D. & Stewart, K. E. Meditation, mindfulness, and attention: a meta-analysis. Mindfulness (N. Y). 3 , 1332–1349 (2021).

Falcone, G. & Jerram, M. Brain activity in mindfulness depends on experience: a meta-analysis of fMRI studies. Mindfulness (N. Y). 9 , 1319–1329 (2018).

Brewer, J. A. et al. Meditation experience is associated with differences in default mode network activity and connectivity. Proc. Natl. Acad. Sci. U. S. A. 108 , 20254–20259 (2011).

Blanck, P. et al. Effects of mindfulness exercises as stand-alone intervention on symptoms of anxiety and depression: systematic review and meta-analysis. Behav. Res. Ther. 102 , 25–35 (2018).

Goldberg, S. B. et al. Mindfulness-based cognitive therapy for the treatment of current depressive symptoms: a meta-analysis. Cogn. Behav. Ther. 48 , 445–462 (2019).

Zhang, D. & Raichle, M. E. Disease and the brain’s dark energy. Nat. Rev. Neurol. 6 , 15–28 (2010).

Doucet, G. E. et al. Transdiagnostic and disease-specific abnormalities in the default-mode network hubs in psychiatric disorders: a meta-analysis of resting-state functional imaging studies. Eur. Psychiatry 63 , 19 (2020).

Smallwood, J. et al. The neural correlates of ongoing conscious thought. iScience 24 , 1–15 (2021).

Calhoun, V. D., Miller, R., Pearlson, G. & Adali, T. The chronnectome: time-varying connectivity networks as the next frontier in fMRI data discovery. Neuron 84 , 262–274 (2014).

Article   CAS   PubMed   PubMed Central   Google Scholar  

Karapanagiotidis, T. et al. The psychological correlates of distinct neural states occurring during wakeful rest. Sci. Rep. 10 , 1–11 (2020).

Kucyi, A. et al. Prediction of stimulus-independent and task-unrelated thought from functional brain networks. Nat. Commun. 12 , 1–17 (2021).

Turnbull, A. et al. Reductions in task positive neural systems occur with the passage of time and are associated with changes in ongoing thought. Sci. Rep. 10 , 1–10 (2020).

CAS   Google Scholar  

Turnbull, A. et al. The ebb and flow of attention: between-subject variation in intrinsic connectivity and cognition associated with the dynamics of ongoing experience. Neuroimage 185 , 286–299 (2019).

Fox, M. C., Ericsson, K. A. & Best, R. Do procedures for verbal reporting of thinking have to be reactive? A meta-analysis and recommendations for best reporting methods. Psychol. Bull. 137 , 316–344 (2011).

Pennebaker, J. W. et al. Levels of thinking. Personal. Soc. Psychol. Bull. 16 , 743–757 (1990).

van Vugt, M. K., Hitchcock, P., Shahar, B. & Britton, W. The effects of mindfulness-based cognitive therapy on affective memory recall dynamics in depression: a mechanistic model of rumination. Front. Hum. Neurosci. 6 , 257 (2012).

PubMed   PubMed Central   Google Scholar  

Download references

Acknowledgements

We are grateful for assistance from Sylvia Zarnescu, Rohith Boyilla, Zain Majeed, and Aakash Unarker with transcription and coding of the think aloud paradigm. This work was supported by the University of Arizona, the Arizona Alzheimer’s Consortium, and the National Institute on Aging.

Author information

Authors and affiliations.

Department of Psychology, University of Arizona, 1503 E University Blvd., Tucson, AZ, 85721, USA

Quentin Raffaelli, Nadia-Anais de Stefano, Matthias R. Mehl, Kate Chambers, Surya A. Fitzgerald, Ramsey Wilcox, Eric S. Andrews, Matthew D. Grilli, Mary-Frances O’Connor & Jessica R. Andrews-Hanna

Department of Psychology, University of New Hampshire, Durham, NH, USA

Caitlin Mills

Department of Psychology, University of British Columbia, Vancouver, BC, Canada

Kalina Christoff

Centre for Brain Health, University of British Columbia, Vancouver, BC, Canada

Department of Neurology, University of Arizona, Tucson, AZ, USA

Matthew D. Grilli

Evelyn F. McKnight Brain Institute, University of Arizona, Tucson, AZ, USA

Matthew D. Grilli & Jessica R. Andrews-Hanna

Cognitive Science, University of Arizona, Tucson, AZ, USA

• Jessica R. Andrews-Hanna

You can also search for this author in PubMed   Google Scholar

Contributions

Q.R., J.R.A.-H., M.-F.O’C., M.R.M., and R.W. conceived of and designed the studies. Q.R. and N.-A.S. collected the data. Q.R., N.-A.S., K.C., and S.A.F. manually coded the data. Q.R. and C.M. analyzed the data. Q.R. and J.R.A.-H. wrote the paper. Q.R., C.M., and J.R.A.-H. prepared the figures. Q.R., J.R.A.-H., C.M., M.R.M., M.-F.O’C., M.D.G., K.C., and E.S.A. discussed the results and provided feedback on the manuscript.

Corresponding authors

Correspondence to Quentin Raffaelli or Jessica R. Andrews-Hanna .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Supplementary information., rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Raffaelli, Q., Mills, C., de Stefano, NA. et al. The think aloud paradigm reveals differences in the content, dynamics and conceptual scope of resting state thought in trait brooding. Sci Rep 11 , 19362 (2021). https://doi.org/10.1038/s41598-021-98138-x

Download citation

Received : 03 May 2021

Accepted : 25 August 2021

Published : 30 September 2021

DOI : https://doi.org/10.1038/s41598-021-98138-x

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

This article is cited by

An empirical study: self-criticism, rumination response style and alexithymia traits in young adults.

• Sanjoni Sethi

Current Psychology (2024)

A dorsomedial prefrontal cortex-based dynamic functional connectivity model of rumination

• Jungwoo Kim
• Choong-Wan Woo

Nature Communications (2023)

• Quentin Raffaelli
• Eric S. Andrews

Scientific Reports (2023)

Recent advances in the neuroscience of spontaneous and off-task thought: implications for mental health

• Aaron Kucyi
• Julia W. Y. Kam
• Susan Whitfield-Gabrieli

Nature Mental Health (2023)

Cluster-based characterization of consistencies in individuals' thought profiles at rest in a cohort of 1779 French university students

• Sandrine Cremona
• Marc Joliot
• Emmanuel Mellet

Current Psychology (2023)

By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

The Use of Think-aloud Methods in Qualitative Research An Introduction to Think-aloud Methods

• Elizabeth Charters

How to Cite

• Endnote/Zotero/Mendeley (RIS)

Developed By

Information.

• For Readers
• For Authors
• For Librarians

Usability Testing Your Writing: Using the Think Aloud Protocol to Get Constructive Reader Feedback

• First Online: 26 April 2023

Cite this chapter

• Michelle Riedlinger 3 &
• Glen Thomas 3  

800 Accesses

Students often have difficulty giving constructive feedback to others on their writing. They can also struggle to take and apply constructive reader feedback. This activity introduces students to the Think Aloud Protocol, a tool for giving and obtaining feedback in a productive way. By the end of this activity, students will be able to confidently give feedback to others on their writing. The activity also helps students reframe how and why feedback from others is useful. It helps students appreciate reader engagement with their written work, as writers can see how reader feedback helps them improve their writing.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

• Available as EPUB and PDF
• Read on any device
• Instant download
• Own it forever
• Durable hardcover edition
• Dispatched in 3 to 5 business days
• Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Website Links

A useful article demonstrating the usefulness of the Think Aloud Protocol for collecting reliable information on thought processes. ( https://psycnet.apa.org/record/2010-23810-001?doi=1 ). Accessed 21 Jan 2021 .

A useful article with some practical examples of useful Think Aloud feedback. ( https://link.springer.com/article/10.1007/s40037-018-0492-z ). Accessed 21 Jan 2021 .

A useful research article on using the Think Aloud Protocol to help writers anticipate readers’ needs. ( https://journals.sagepub.com/doi/pdf/10.1177/0741088392009002001 ). Accessed 21 Jan 2021 .

Download references

Author information

Authors and affiliations.

School of Communication, Queensland University of Technology, Kelvin Grove, QLD, Australia

Michelle Riedlinger & Glen Thomas

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Michelle Riedlinger .

Editor information

Editors and affiliations.

School of Chemical and Molecular Biosciences, The University of Queensland, Brisbane, QLD, Australia

Susan Rowland

School of Biological Sciences, The University of Queensland, Brisbane, QLD, Australia

Louise Kuchel

1 Electronic Supplementary Material

. Think Aloud Protocol Slides (Riedlinger and Thomas) (PPTX 612 kb)

Rights and permissions

Reprints and permissions

Copyright information

© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG

About this chapter

Riedlinger, M., Thomas, G. (2023). Usability Testing Your Writing: Using the Think Aloud Protocol to Get Constructive Reader Feedback. In: Rowland, S., Kuchel, L. (eds) Teaching Science Students to Communicate: A Practical Guide. Springer, Cham. https://doi.org/10.1007/978-3-030-91628-2_25

Download citation

DOI : https://doi.org/10.1007/978-3-030-91628-2_25

Published : 26 April 2023

Publisher Name : Springer, Cham

Print ISBN : 978-3-030-91627-5

Online ISBN : 978-3-030-91628-2

eBook Packages : Biomedical and Life Sciences Biomedical and Life Sciences (R0)

Share this chapter

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Publish with us

Policies and ethics

• Find a journal
• Track your research

Cookies on GOV.UK

We use some essential cookies to make this website work.

We’d like to set additional cookies to understand how you use GOV.UK, remember your settings and improve government services.

We also use cookies set by other sites to help us deliver content from their services.

You have accepted additional cookies. You can change your cookie settings at any time.

You have rejected additional cookies. You can change your cookie settings at any time.

• Health and social care
• Public health
• Health improvement

Think aloud study: qualitative studies

How to use a think aloud study to evaluate your digital health product.

Think aloud methods ask participants to verbalise what they are thinking and doing as they perform a task using your digital health product or service.

What to use it for

It is carried out to understand user experience of your product, for example, to uncover usability problems with navigation, or highlight content that could be improved. It is usually used as a method for carrying out usability testing .

You can use think aloud:

• when you are developing your product, to work out how to make it better (formative or iterative evaluation)
• once you have a product and want to improve it, for example by adding new content or features

Benefits of think aloud include:

• it provides insight into people’s experiences as they interact with your product
• it can be cheap to carry out
• it can be carried out with low numbers of participants

Drawbacks of think aloud include:

• it relies on people verbalising thoughts and impressions, rather than objective measures
• participants may say what they believe to be the right answer rather than what they really think (social desirability). This can distort your results and conclusions
• it cannot tell you whether your product has an impact on the outcomes you wanted

How to carry out a think aloud study

Plan the study.

Decide what questions you want your study to answer. For example, whether users can find particular content or what their understanding is of the information presented.

Create a written plan, including:

• tasks for the user to complete while using your digital product
• a set of open-ended questions

Decide how many participants you want to recruit and how long you want the sessions to last (45 to 90 minutes works well).

Conduct the study

Typically, you should have a facilitator who runs the study and observers to make notes. Having more than one researcher making notes can increase the reliability of the findings. This is especially important if you’re not planning to record and transcribe the sessions.

Ask the participants to complete the tasks you have planned. This should be uninterrupted as far as possible, although the moderator will probably need to give some prompts. Explain to the participants how a think aloud works: they should tell you their thoughts, reactions and emotions as they occur while they are performing the task (concurrent think-aloud).

Another possible approach is a retrospective think aloud. Here, the participant talks through an interaction they previously performed, often using a video recording of the interaction as a prompt. This avoids the act of thinking aloud from interfering with the performance of the task, but the participant may not remember as much.

Before the think aloud, it may also help to explain to participants that:

• this is not a test and there are no right or wrong answers; it’s about understanding how the product or service works for them
• they can be critical about the product – the researchers won’t be offended

Participants don’t need to provide an explanation of their thoughts. They may feel they need to justify their thoughts, especially in front of a researcher who we may have never met, but there are various ways to overcome this obstacle in a think aloud session. For example:

• you could use a warm-up task to help the participant to focus. For example, you could start the think aloud with another section of your website that isn’t the focus of your research.
• the researcher could sit behind the participant to decrease distraction

If the participant falls silent, prompt them to verbalise their thoughts – for example, ‘what are you thinking?’. Try not to do this too often or to lead the participant towards certain answers.

The qualitative data from a think aloud study is classified into meaningful categories. You can do this using various analysis methods typically used for qualitative studies, such as:

• thematic analysis
• framework analysis
• basic counting of numerical data, such as categories or frequencies

Example: Evaluating digital health products - as we expand, we test the structure and content

While developing this resource, the one you are reading now, we created more content and added new pages. As the resource grew, the structure and navigation of the website became less fit for purpose and we decided to conduct a think aloud study to investigate how it could be improved.

We gave participants a scenario: imagine you are developers of an app designed to help insomnia and you want to evaluate it. We asked them to say everything they thought of out loud as they used the resource to complete the task. We showed participants 3 versions of the website homepage: the existing version, and two alternative prototypes. We alternated the order in which participants saw the different versions.

We conducted 4 think aloud sessions, with a facilitator and 2 observers taking comprehensive notes. We collated the findings using a collaborative online tool, grouping observations into:

• what aspects helped participants
• what didn’t help participants
• any additional features they needed

Each category had a different colour on our board (green – helpful, pink – unhelpful, orange – new feature wanted). We organised them and looked at similarities in two new versions, and differences between versions 1 and 2.

Some findings were the same across all participants:

• adding numbered sections in the guidance helped them to navigate the information
• more bullet points helped them to scan the content
• subheadings helped to break up long lists of content and make them less overwhelming

Based on the feedback, we created a version of the website structure, which can be validated with a second set of user testing.

More information and resources

Participants may find it difficult and unnatural to say everything that comes to their mind out loud. Here is a video by the Nielsen Norman Group explaining what it means to ‘think aloud’ . The resource also explains how to create your own demo video for your study.

GOV.UK guidance on using think aloud as part of moderated usability testing

GOV.UK user research blog with practical tips for running a think aloud study

Examples of think aloud studies in digital health

McCall and others (2021), Development of a Mobile App to Support Self-Management of Anxiety and Depression in African American Women: A Usability Study . Researchers used think aloud as one of their methods to evaluate the usability of the app prototype tailored to the needs of African American women to help with anxiety and depression.

Szinay and others (2021), Influences on the Uptake of Health and Well-being Apps and Curated App Portals: Think-Aloud and Interview Study . Researchers were interested in how people choose apps for health and wellbeing. They used think aloud to explore what influences the selection process when looking online.

Related content

Is this page useful.

• Yes this page is useful
• No this page is not useful

Help us improve GOV.UK

Don’t include personal or financial information like your National Insurance number or credit card details.

To help us improve GOV.UK, we’d like to know more about your visit today. We’ll send you a link to a feedback form. It will take only 2 minutes to fill in. Don’t worry we won’t send you spam or share your email address with anyone.

Skip navigation

World Leaders in Research-Based User Experience

Thinking aloud: the #1 usability tool.

January 15, 2012 2012-01-15

• Email article
• Share on LinkedIn
• Share on Twitter

"Thinking aloud may be the single most valuable usability engineering method."

I wrote this in my 1993 book, Usability Engineering , and I stand by this assessment today. The fact that the same method has remained #1 for 19 years is a good indication of the longevity of usability methods .

Usability guidelines live for a long time ; usability methods live even longer. Human behavior changes much more slowly than the technology we all find so fascinating, and the best approaches to studying this behavior hardly change at all.

In This Article:

Defining thinking aloud testing, benefits of think-aloud, downsides of think-aloud.

To define thinking aloud , I'll paraphrase what I said 19 years ago:

Definition: In a thinking aloud test, you ask test participants to use the system while continuously thinking out loud — that is, simply verbalizing their thoughts as they move through the user interface.

("Simply" ought to be in quotes, because it's not that simple for most people to keep up a running monologue. The test facilitator typically has to prompt users to keep them talking.)

To run a basic thinking aloud usability study, you need to do only 3 things:

• Recruit representative users .
• Give them representative tasks to perform.
• Shut up and let the users do the talking.

The method has a host of advantages. Most important, it serves as a window on the soul , letting you discover what users really think about your design. In particular, you hear their misconceptions, which usually turn into actionable redesign recommendations: when users misinterpret design elements, you need to change them. Even better, you usually learn why users guess wrong about some parts of the UI and why they find others easy to use.

The thinking aloud method also offers the benefits of being:

• Cheap. No special equipment is needed; you simply sit next to a user and take notes as he or she talks. It takes about a day to collect data from a handful of users , which is all that's needed for the most important insights.
• Robust. Most people are poor facilitators and don't run the study exactly according to the proper methodology. But, unless you blatantly bias users by putting words into their mouths, you'll still get reasonably good findings, even from a poorly run study. In contrast, quantitative (statistical) usability studies are ripe with methodology problems  and the smallest mistake can doom a study and make the findings directly misleading. Quant studies are also much more expensive .
• Flexible. You can use the method at any stage in the development lifecycle, from early paper prototypes  to fully implemented, running systems. Thinking aloud is particularly suited for Agile projects . You can use this method to evaluate any type of user interface with any form of technology (although it's a bit tricky to use thinking aloud with speech interfaces — see report on How to Conduct Usability Evaluations for Accessibility  for advice on testing with blind or low-vision users who rely on screen readers such as JAWS). Websites, software applications, intranets, consumer products, enterprise software, mobile design: doesn't matter — thinking aloud addresses them all, because we rely on the users doing the thinking.
• Convincing. The most hard-boiled developers, arrogant designers, and tight-fisted executives usually soften up when they get direct exposure to how customers think about their work. Getting the rest of your team (and management) to sit in on a few thinking-aloud sessions doesn't take a lot of their time and is the best way to motivate them to pay attention to usability. (For more on how to motivate teams to deliver superior user experiences, see the UX Basic Training  course.)
• Easy to learn. We teach the basics in a day  and provide thorough team training in a 2-day "Learning-by-Doing" course . Of course, this doesn't cover all the twists and advanced modifications needed to hang out your shingle as a usability consultant, but the point is that you don't need these extras to run basic tests for your own design team .

Being cheap and robust are huge upsides of qualitative methods such as thinking aloud. But the flip side is that the method doesn't lend itself to detailed statistics , unless you run a huge, expensive study. You can certainly do this — I simply don't recommend it for the vast majority of projects. Better to conserve your budget and invest in more design iterations .

Other problems:

• Unnatural situation. Unless they're a bit weird, most people don't sit and talk to themselves all day. This makes it hard for test participants to keep up the required monologue. Luckily, users are typically quite willing to try their best, and they quickly become so engaged in the test tasks  that they all but forget that they're in a study. You can show users a short video demo of a think-aloud session to quickly and vividly explain what's expected of them.
• Filtered statements (vs. brain dump). Users are supposed to say things as soon as they come to mind rather than reflect on their experience and provide an edited commentary after the fact. However, most people want to appear smart, and thus there's a risk that they won't speak until they've thought through the situation in detail. Don't fall for this trap: it's essential to get the user's raw stream of thought. Typically, you have to prompt users to keep them talking.
• Biasing user behavior. Prompts and clarifying questions are usually necessary, but from an untrained facilitator, such interruptions can very easily change user behavior. In such cases, the resulting behavior doesn't represent real use, so you can't base design decisions on the outcome. At the very least, try to identify those cases where you've biased the user so you can discard that small part of the study. (It's worse when you don't know that you've done wrong — then you risk giving the design team bad advice.)
• No panacea. That this one method isn't the only usability tool you'll ever need is not a true downside, as long as you are willing to use other methods from time to time. Thinking aloud serves many purposes, but not all purposes. Once you get a few years' experience with usability, you'll want to use a wider range of user research methods .

Don't let the downsides get you down. If you haven't tried it before, go run a quick thinking aloud study on your current design project right now. Because these simplified studies are so cheap, weekly user testing  is completely feasible — so if you make a few mistakes the first time, you can always correct them next week.

Related Courses

Usability testing.

Learn how to plan, conduct, and analyze your own studies, whether in person or remote

Lean UX and Agile

Applying Lean UX approaches for Agile environments

Remote User Research

Collect insights without leaving your desk

Related Topics

• User Testing User Testing

Learn More:

Please accept marketing cookies to view the embedded video. https://www.youtube.com/watch?v=tXJqAYDbRzI

Help Users Think Aloud

Data vs. Findings vs. Insights

Sara Ramaswamy · 3 min

Usability Test Facilitation: 6 Mistakes to Avoid

Kate Moran and Maria Rosala · 6 min

Introducing a Participant to a Usability Test: A Demonstration

Maria Rosala and Kate Moran · 4 min

Related Articles:

Qualitative Usability Testing: Study Guide

Kate Moran · 5 min

Group Notetaking for User Research

Susan Farrell · 10 min

Project Management for User Research: The Plan

Susan Farrell · 7 min

Employees as Usability-Test Participants

Angie Li · 5 min

Team Members Behaving Badly During Usability Tests

Hoa Loranger · 7 min

Time Budgets for Usability Sessions

Jakob Nielsen · 4 min

• Open access
• Published: 02 May 2024

Use of the International IFOMPT Cervical Framework to inform clinical reasoning in postgraduate level physiotherapy students: a qualitative study using think aloud methodology

• Katie L. Kowalski 1 ,
• Heather Gillis 1 ,
• Katherine Henning 1 ,
• Paul Parikh 1 ,
• Jackie Sadi 1 &
• Alison Rushton 1  

BMC Medical Education volume  24 , Article number:  486 ( 2024 ) Cite this article

185 Accesses

Metrics details

Vascular pathologies of the head and neck are rare but can present as musculoskeletal problems. The International Federation of Orthopedic Manipulative Physical Therapists (IFOMPT) Cervical Framework (Framework) aims to assist evidence-based clinical reasoning for safe assessment and management of the cervical spine considering potential for vascular pathology. Clinical reasoning is critical to physiotherapy, and developing high-level clinical reasoning is a priority for postgraduate (post-licensure) educational programs.

To explore the influence of the Framework on clinical reasoning processes in postgraduate physiotherapy students.

Qualitative case study design using think aloud methodology and interpretive description, informed by COnsolidated criteria for REporting Qualitative research. Participants were postgraduate musculoskeletal physiotherapy students who learned about the Framework through standardized delivery. Two cervical spine cases explored clinical reasoning processes. Coding and analysis of transcripts were guided by Elstein’s diagnostic reasoning components and the Postgraduate Musculoskeletal Physiotherapy Practice model. Data were analyzed using thematic analysis (inductive and deductive) for individuals and then across participants, enabling analysis of key steps in clinical reasoning processes and use of the Framework. Trustworthiness was enhanced with multiple strategies (e.g., second researcher challenged codes).

For all participants ( n  = 8), the Framework supported clinical reasoning using primarily hypothetico-deductive processes. It informed vascular hypothesis generation in the patient history and testing the vascular hypothesis through patient history questions and selection of physical examination tests, to inform clarity and support for diagnosis and management. Most participant’s clinical reasoning processes were characterized by high-level features (e.g., prioritization), however there was a continuum of proficiency. Clinical reasoning processes were informed by deep knowledge of the Framework integrated with a breadth of wider knowledge and supported by a range of personal characteristics (e.g., reflection).

Conclusions

Findings support use of the Framework as an educational resource in postgraduate physiotherapy programs to inform clinical reasoning processes for safe and effective assessment and management of cervical spine presentations considering potential for vascular pathology. Individualized approaches may be required to support students, owing to a continuum of clinical reasoning proficiency. Future research is required to explore use of the Framework to inform clinical reasoning processes in learners at different levels.

Peer Review reports

Introduction

Musculoskeletal neck pain and headache are highly prevalent and among the most disabling conditions globally that require effective rehabilitation [ 1 , 2 , 3 , 4 ]. A range of rehabilitation professionals, including physiotherapists, assess and manage musculoskeletal neck pain and headache. Assessment of the cervical spine can be a complex process. Patients can present to physiotherapy with vascular pathology masquerading as musculoskeletal pain and dysfunction, as neck pain and/or headache as a common first symptom [ 5 ]. While vascular pathologies of the head and neck are rare [ 6 ], they are important considerations within a cervical spine assessment to facilitate the best possible patient outcomes [ 7 ]. The International IFOMPT (International Federation of Orthopedic Manipulative Physical Therapists) Cervical Framework (Framework) provides guidance in the assessment and management of the cervical spine region, considering the potential for vascular pathologies of the neck and head [ 8 ]. Two separate, but related, risks are considered: risk of misdiagnosis of an existing vascular pathology and risk of serious adverse event following musculoskeletal interventions [ 8 ].

The Framework is a consensus document iteratively developed through rigorous methods and the best contemporary evidence [ 8 ], and is also published as a Position Statement [ 7 ]. Central to the Framework are clinical reasoning and evidence-based practice, providing guidance in the assessment of the cervical spine region, considering the potential for vascular pathologies in advance of planned interventions [ 7 , 8 ]. The Framework was developed and published to be a resource for practicing musculoskeletal clinicians and educators. It has been implemented widely within IFOMPT postgraduate (post-licensure) educational programs, influencing curricula by enabling a comprehensive and systemic approach when considering the potential for vascular pathology [ 9 ]. Frequently reported curricula changes include an emphasis on the patient history and incorporating Framework recommended physical examination tests to evaluate a vascular hypothesis [ 9 ]. The Framework aims to assist musculoskeletal clinicians in their clinical reasoning processes, however no study has investigated students’ use of the Framework to inform their clinical reasoning.

Clinical reasoning is a critical component to physiotherapy practice as it is fundamental to assessment and diagnosis, enabling physiotherapists to provide safe and effective patient-centered care [ 10 ]. This is particularly important for postgraduate physiotherapy educational programs, where developing a high level of clinical reasoning is a priority for educational curricula [ 11 ] and critical for achieving advanced practice physiotherapy competency [ 12 , 13 , 14 , 15 ]. At this level of physiotherapy, diagnostic reasoning is emphasized as an important component of a high level of clinical reasoning, informed by advanced use of domain-specific knowledge (e.g., propositional, experiential) and supported by a range of personal characteristics (e.g., adaptability, reflective) [ 12 ]. Facilitating the development of clinical reasoning improves physiotherapist’s performance and patient outcomes [ 16 ], underscoring the importance of clinical reasoning to physiotherapy practice. Understanding students’ use of the Framework to inform their clinical reasoning can support optimal implementation of the Framework within educational programs to facilitate safe and effective assessment and management of the cervical spine for patients.

To explore the influence of the Framework on the clinical reasoning processes in postgraduate level physiotherapy students.

Using a qualitative case study design, think aloud case analyses enabled exploration of clinical reasoning processes in postgraduate physiotherapy students. Case study design allows evaluation of experiences in practice, providing knowledge and accounts of practical actions in a specific context [ 17 ]. Case studies offer opportunity to generate situationally dependent understandings of accounts of clinical practice, highlighting the action and interaction that underscore the complexity of clinical decision-making in practice [ 17 ]. This study was informed by an interpretive description methodological approach with thematic analysis [ 18 , 19 ]. Interpretive description is coherent with mixed methods research and pragmatic orientations [ 20 , 21 ], and enables generation of evidence-based disciplinary knowledge and clinical understanding to inform practice [ 18 , 19 , 22 ]. Interpretive description has evolved for use in educational research to generate knowledge of educational experiences and the complexities of health care education to support achievement of educational objectives and professional practice standards [ 23 ]. The COnsolidated criteria for REporting Qualitative research (COREQ) informed the design and reporting of this study [ 24 ].

Research team

All research team members hold physiotherapy qualifications, and most hold advanced qualifications specializing in musculoskeletal physiotherapy. The research team is based in Canada and has varying levels of academic credentials (ranging from Clinical Masters to PhD or equivalent) and occupations (ranging from PhD student to Director of Physical Therapy). The final author (AR) is also an author of the Framework, which represents international and multiprofessional consensus. Authors HG and JS are lecturers on one of the postgraduate programs which students were recruited from. The primary researcher and first author (KK) is a US-trained Physical Therapist and Postdoctoral Research Associate investigating spinal pain and clinical reasoning in the School of Physical Therapy at Western University. Authors KK, KH and PP had no prior relationship with the postgraduate educational programs, students, or the Framework.

Study setting

Western University in London, Ontario, Canada offers a one-year Advanced Health Care Practice (AHCP) postgraduate IFOMPT-approved Comprehensive Musculoskeletal Physiotherapy program (CMP) and a postgraduate Sport and Exercise Medicine (SEM) program. Think aloud case analyses interviews were conducted using Zoom, a viable option for qualitative data collection and audio-video recording of interviews that enables participation for students who live in geographically dispersed areas across Canada [ 25 ]. Interviews with individual participants were conducted by one researcher (KK or KH) in a calm and quiet environment to minimize disruption to the process of thinking aloud [ 26 ].

Participants

AHCP postgraduate musculoskeletal physiotherapy students ≥ 18 years of age in the CMP and SEM programs were recruited via email and an introduction to the research study during class by KK, using purposive sampling to ensure theoretical representation. The purposive sample ensured key characteristics of participants were included, specifically gender, ethnicity, and physiotherapy experience (years, type). AHCP students must have attended standardized teaching about the Framework to be eligible to participate. Exclusion criteria included inability to communicate fluently in English. As think-aloud methodology seeks rich, in-depth data from a small sample [ 27 ], this study sought to recruit 8–10 AHCP students. This range was informed by prior think aloud literature and anticipated to balance diversity of participant characteristics, similarities in musculoskeletal physiotherapy domain knowledge and rich data supporting individual clinical reasoning processes [ 27 , 28 ].

Learning about the IFOMPT Cervical Framework

CMP and SEM programs included standardized teaching of the Framework to inform AHCP students’ clinical reasoning in practice. Delivery included a presentation explaining the Framework, access to the full Framework document [ 8 ], and discussion of its role to inform practice, including a case analysis of a cervical spine clinical presentation, by research team members AR and JS. The full Framework document that is publicly available through IFOMPT [ 8 ] was provided to AHCP students as the Framework Position Statement [ 7 ] was not yet published. Discussion and case analysis was led by AHCP program leads in November 2021 (CMP, including research team member JS) and January 2022 (SEM).

Think aloud case analyses data collection

Using think aloud methodology, the analytical processes of how participants use the Framework to inform clinical reasoning were explored in an interview with one research team member not involved in AHCP educational programs (KK or KH). The think aloud method enables description and explanation of complex information paralleling the clinical reasoning process and has been used previously in musculoskeletal physiotherapy [ 29 , 30 ]. It facilitates the generation of rich verbal [ 27 ]as participants verbalize their clinical reasoning protocols [ 27 , 31 ]. Participants were aware of the aim of the research study and the research team’s clinical and research backgrounds, supporting an open environment for depth of data collection [ 32 ]. There was no prior relationship between participants and research team members conducting interviews.

Participants were instructed to think aloud their analysis of two clinical cases, presented in random order (Supplementary  1 ). Case information was provided in stages to reflect the chronology of assessment of patients in practice (patient history, planning the physical examination, physical examination, treatment). Use of the Framework to inform clinical reasoning was discussed at each stage. The cases enabled participants to identify and discuss features of possible vascular pathology, treatment indications and contraindications/precautions, etc. Two research study team members (HG, PP) developed cases designed to facilitate and elicit clinical reasoning processes in neck and head pain presentations. Cases were tested against the research team to ensure face validity. Cases and think aloud prompts were piloted prior to use with three physiotherapists at varying levels of practice to ensure they were fit for purpose.

Data collection took place from March 30-August 15, 2022, during the final terms of the AHCP programs and an average of 5 months after standardized teaching about the Framework. During case analysis interviews, participants were instructed to constantly think aloud, and if a pause in verbalizations was sustained, they were reminded to “keep thinking aloud” [ 27 ]. As needed, prompts were given to elicit verbalization of participants’ reasoning processes, including use of the Framework to inform their clinical reasoning at each stage of case analysis (Supplementary  2 ). Aside from this, all interactions between participants and researchers minimized to not interfere with the participant’s thought processes [ 27 , 31 ]. When analysis of the first case was complete, the researcher provided the second case, each lasting 35–45 min. A break between cases was offered. During and after interviews, field notes were recorded about initial impressions of the data collection session and potential patterns appearing to emerge [ 33 ].

Data analysis

Data from think aloud interviews were analyzed using thematic analysis [ 30 , 34 ], facilitating identification and analysis of patterns in data and key steps in the clinical reasoning process, including use of the Framework to enable its characterization (Fig.  1 ). As established models of clinical reasoning exist, a hybrid approach to thematic analysis was employed, incorporating inductive and deductive processes [ 35 ], which proceeded according to 5 iterative steps: [ 34 ]

Data analysis steps

Familiarize with data: Audio-visual recordings were transcribed verbatim by a physiotherapist external to the research team. All transcripts were read and re-read several times by one researcher (KK), checking for accuracy by reviewing recordings as required. Field notes supported depth of familiarization with data.

Generate initial codes: Line-by-line coding of transcripts by one researcher (KK) supported generation of initial codes that represented components, patterns and meaning in clinical reasoning processes and use of the Framework. Established preliminary coding models were used as a guide. Elstein’s diagnostic reasoning model [ 36 ] guided generating initial codes of key steps in clinical reasoning processes (Table  1 a) [ 29 , 36 ]. Leveraging richness of data, further codes were generated guided by the Postgraduate Musculoskeletal Physiotherapy Practice model, which describes masters level clinical practice (Table  1 b) [ 12 ]. Codes were refined as data analysis proceeded. All codes were collated within participants along with supporting data.

Generate initial themes within participants: Coded data was inductively grouped into initial themes within each participant, reflecting individual clinical reasoning processes and use of the Framework. This inductive stage enabled a systematic, flexible approach to describe each participant’s unique thinking path, offering insight into the complexities of their clinical reasoning processes. It also provided a comprehensive understanding of the Framework informing clinical reasoning and a rich characterization of its components, aiding the development of robust, nuanced insights [ 35 , 37 , 38 ]. Initial themes were repeatedly revised to ensure they were grounded in and reflected raw data.

Develop, review and refine themes across participants: Initial themes were synthesized across participants to develop themes that represented all participants. Themes were reviewed and refined, returning to initial themes and codes at the individual participant level as needed.

Organize themes into established models: Themes were deductively organized into established clinical reasoning models; first into Elstein’s diagnostic reasoning model, second into the Postgraduate Musculoskeletal Physiotherapy Practice model to characterize themes within each diagnostic reasoning component [ 12 , 36 ].

Trustworthiness of findings

The research study was conducted according to an a priori protocol and additional steps were taken to establish trustworthiness of findings [ 39 ]. Field notes supported deep familiarization with data and served as a means of data source triangulation during analysis [ 40 ]. One researcher coded transcripts and a second researcher challenged codes, with codes and themes rigorously and iteratively reviewed and refined. Frequent debriefing sessions with the research team, reflexive discussions with other researchers and peer scrutiny of initial findings enabled wider perspectives and experiences to shape analysis and interpretation of findings. Several strategies were implemented to minimize the influence of prior relationships between participants and researchers, including author KK recruiting participants, KK and KH collecting/analyzing data, and AR, JS, HG and PP providing input on de-identified data at the stage of synthesis and interpretation.

Nine AHCP postgraduate level students were recruited and participated in data collection. One participant was withdrawn because of unfamiliarity with the standardized teaching session about use of the Framework (no recall of session), despite confirmation of attendance. Data from eight participants were used for analysis (CMP: n  = 6; SEM: n  = 2; Table  2 ), which achieved sample size requirements for think aloud methodology of rich and in-depth data [ 27 , 28 ].

Diagnostic reasoning components

Informed by the Framework, all components of Elstein’s diagnostic reasoning processes [ 36 ] were used by participants, including use of treatment with physiotherapy interventions to aid diagnostic reasoning. An illustrative example is presented in Supplement  3 . Clinical reasoning used primarily hypothetico-deductive processes reflecting a continuum of proficiency, was informed by deep Framework knowledge and breadth of prior knowledge (e.g., experiential), and supported by a range of personal characteristics (e.g., justification for decisions).

Cue acquisition

All participants sought to acquire additional cues early in the patient history, and for some this persisted into the medical history and physical examination. Cue acquisition enabled depth and breadth of understanding patient history information to generate hypotheses and factors contributing to the patient’s pain experience (Table  3 ). All participants asked further questions to understand details of the patients’ pain and their presentation, while some also explored the impact of pain on patient functioning and treatments received to date. There was a high degree of specificity to questions for most participants. Ongoing clinical reasoning processes through a thorough and complete assessment, even if the patient had previously received treatment for similar symptoms, was important for some participants. Cue acquisition was supported by personal characteristics including a patient-centered approach (e.g., understanding the patient’s beliefs about pain) and one participant reflected on their approach to acquiring patient history cues.

Hypothesis generation

Participants generated an average of 4.5 hypotheses per case (range: 2–8) and most hypotheses (77%) were generated rapidly early in the patient history. Knowledge from the Framework about patient history features of vascular pathology informed vascular hypothesis generation in the patient history for all participants in both cases (Table  4 ). Vascular hypotheses were also generated during the past medical history, where risk factors for vascular pathology were identified and interpreted by some participants who had high levels of suspicion for cervical articular involvement. Non-vascular hypotheses were generated during the physical examination by some participants to explain individual physical examination or patient history cues. Deep knowledge of the patient history section in the Framework supported high level of cue identification and interpretation for generating vascular hypotheses. Initial hypotheses were prioritized by some participants, however the level of specificity of hypotheses varied.

Cue evaluation

All participants evaluated cues throughout the patient history and physical examination in relationship to hypotheses generated, indicating use of hypothetico-deductive reasoning processes (Table  5 ). Framework knowledge of patient history features of vascular pathology was used to test vascular hypotheses and aid differential diagnosis. The patient history section supported high level of cue identification and interpretation of patient history features for all but one participant, and generation of further patient history questions for all participants. The level of specificity of these questions was high for all but one participant. Framework knowledge of recommended physical examination tests, including removal of positional testing, supported planning a focused and prioritized physical examination to further test vascular hypotheses for all participants. No participant indicated intention to use positional testing as part of their physical examination. Treatment with physiotherapy interventions served as a form of cue evaluation, and cues were evaluated to inform prognosis for some participants. At times during the physical examination, some participants demonstrated occasional errors or difficulty with cue evaluation by omitting key physical exam tests (e.g., no cranial nerve assessment despite concerns for trigeminal nerve involvement), selecting physical exam tests in advance of hypothesis generation (e.g., cervical spine instability testing), difficulty interpreting cues, or late selection of a physical examination test. Cue acquisition was supported by a range of personal characteristics. Most participants justified selection of physical examination tests, and some self-reflected on their ability to collect useful physical examination information to inform selection of tests. Precaution to the physical examination was identified by all participants but one, which contributed to an adaptable approach, prioritizing patient safety and comfort. Critical analysis of physical examination information aided interpretation within the context of the patient for most participants.

Hypothesis evaluation

All participants used the Framework to evaluate their hypotheses throughout the patient history and physical examination, continuously shifting their level of support for hypotheses (Table  6 , Supplement  4 ). This informed clarity in the overall level of suspicion for vascular pathology or musculoskeletal diagnoses, which were specific for most participants. Response to treatment with physiotherapy interventions served as a form of hypothesis evaluation for most participants who had low level suspicion for vascular pathology, highlighting ongoing reasoning processes. Hypotheses evaluated were prioritized by ranking according to level of suspicion by some participants. Difficulties weighing patient history and physical examination cues to inform judgement on overall level of suspicion for vascular pathology was demonstrated by some participants who reported that incomplete physical examination data and not being able to see the patient contributed to difficulties. Hypothesis evaluation was supported by the personal characteristic of reflection, where some students reflected on the Framework’s emphasis on the patient history to evaluate a vascular hypothesis.

The Framework supported all participants in clinical reasoning related to treatment (Table  7 ). Treatment decisions were always linked to the participant’s overall level of suspicion for vascular pathology or musculoskeletal diagnosis. Framework knowledge supported participants with high level of suspicion for vascular pathology to refer for further investigations. Participants with a musculoskeletal diagnosis kept the patient for physiotherapy interventions. The Framework patient history section supported patient education about symptoms of vascular pathology and safety netting for some participants. Framework knowledge influenced informed consent processes and risk-benefit analysis to support the selection of musculoskeletal physiotherapy interventions, which were specific and prioritized for some participants. Less Framework knowledge related to treatment was demonstrated by some students, generating unclear recommendations regarding the urgency of referral and use of the Framework to inform musculoskeletal physiotherapy interventions. Treatment was supported by a range of personal characteristics. An adaptable approach that prioritized patient safety and was supported by justification was demonstrated in all participants except one. Shared decision-making enabled the selection of physiotherapy interventions, which were patient-centered (individualized, considered whole person, identified future risk for vascular pathology). Communication with the patient’s family doctor facilitated collaborative patient-centered care for most participants.

This is the first study to explore the influence of the Framework on clinical reasoning processes in postgraduate physiotherapy students. The Framework supported clinical reasoning that used primarily hypothetico-deductive processes. The Framework informed vascular hypothesis generation in the patient history and testing the vascular hypothesis through patient history questions and selection of physical examination tests to inform clarity and support for diagnosis and management. Most postgraduate students’ clinical reasoning processes were characterized by high-level features (e.g. specificity, prioritization). However, some demonstrated occasional difficulties or errors, reflecting a continuum of clinical reasoning proficiency. Clinical reasoning processes were informed by deep knowledge of the Framework integrated with a breadth of wider knowledge and supported by a range of personal characteristics (e.g., justification for decisions, reflection).

Use of the Framework to inform clinical reasoning processes

The Framework provided a structured and comprehensive approach to support postgraduate students’ clinical reasoning processes in assessment and management of the cervical spine region, considering the potential for vascular pathology. Patient history and physical examination information was evaluated to inform clarity and support the decision to refer for further vascular investigations or proceed with musculoskeletal physiotherapy diagnosis/interventions. The Framework is not intended to lead to a vascular pathology diagnosis [ 7 , 8 ], and following the Framework does not guarantee vascular pathologies will be identified [ 41 ]. Rather, it aims to support a process of clinical reasoning to elicit and interpret appropriate patient history and physical examination information to estimate the probability of vascular pathology and inform judgement about the need to refer for further investigations [ 7 , 8 , 42 ]. Results of this study suggest the Framework has achieved this aim for postgraduate physiotherapy students.

The Framework supported postgraduate students in using primarily hypothetico-deductive diagnostic reasoning processes. This is expected given the diversity of vascular pathology clinical presentations precluding a definite clinical pattern and inherent complexity as a potential masquerader of a musculoskeletal problem [ 7 ]. It is also consistent with prior research investigating clinical reasoning processes in musculoskeletal physiotherapy postgraduate students [ 12 ] and clinical experts [ 29 ] where hypothetico-deductive and pattern recognition diagnostic reasoning are employed according to the demands of the clinical situation [ 10 ]. Diagnostic reasoning of most postgraduate students in this study demonstrated features suggestive of high-level clinical reasoning in musculoskeletal physiotherapy [ 12 ], including ongoing reasoning with high-level cue identification and interpretation, specificity and prioritization during assessment and treatment, use of physiotherapy interventions to aid diagnostic reasoning, and prognosis determination [ 12 , 29 , 43 ]. Expert physiotherapy practice has been further described as using a dialectical model of clinical reasoning with seamless transitions between clinical reasoning strategies [ 44 ]. While diagnostic reasoning was a focus in this study, postgraduate students considered a breadth of information as important to their reasoning (e.g., patient’s perspectives of the reason for their pain). This suggests wider reasoning strategies (e.g., narrative, collaborative) were employed to enable shared decision-making within the context of patient-centered care.

Study findings also highlighted a continuum of proficiency in use of the Framework to inform clinical reasoning processes. Not all students demonstrated all characteristics of high-level clinical reasoning and there are suggestions of incomplete reasoning processes, for example occasional errors in evaluating cues. Some students offered explanations such as incomplete case information as factors contributing to difficulties with clinical reasoning processes. However, the ability to critically evaluate incomplete and potentially conflicting clinical information is consistently identified as an advanced clinical practice competency [ 14 , 43 ]. A continuum of proficiency in clinical reasoning in musculoskeletal physiotherapy is supported by wider healthcare professions describing acquisition and application of clinical knowledge and skills as a developmental continuum of clinical competence progressing from novice to expert [ 45 , 46 ]. The range of years of clinical practice experience in this cohort of students (3–14 years) or prior completed postgraduate education may have contributed to the continuum of proficiency, as high-quality and diverse experiential learning is essential for the development of high-level clinical reasoning [ 14 , 47 ].

Deep knowledge of the Framework informs clinical reasoning processes

Postgraduate students demonstrated deep Framework knowledge to inform clinical reasoning processes. All students demonstrated knowledge of patient history features of vascular pathology, recommended physical examination tests to test a vascular hypothesis, and the need to refer if there is a high level of suspicion for vascular pathology. A key development in the recent Framework update is the removal of the recommendation to perform positional testing [ 8 ]. All students demonstrated knowledge of this development, and none wanted to test a vascular hypothesis with positional testing. Most also demonstrated Framework knowledge about considerations for planning treatment with physiotherapy interventions (e.g., risk-benefit analysis, informed consent), though not all, which underscores the continuum of proficiency in postgraduate students. Rich organization of multidimensional knowledge is a required component for high level clinical reasoning and is characteristic of expert physiotherapy practice [ 10 , 48 , 49 ]. Most postgraduate physiotherapy students displayed this expert practice characteristic through integration of deep Framework knowledge with a breadth of prior knowledge (e.g., experiential, propositional) to inform clinical reasoning processes. This highlights the utility of the Framework in postgraduate physiotherapy education to develop advanced level evidence-based knowledge informing clinical reasoning processes for safe assessment and management of the cervical spine, considering the potential for vascular pathology [ 9 , 8 , 50 , 51 , 52 ].

Framework supports personal characteristics to facilitate integration of knowledge and clinical reasoning

The Framework supported personal characteristics of postgraduate students, which are key drivers for the complex integration of advanced knowledge and high-level clinical reasoning [ 10 , 12 , 48 ]. For all students, the Framework supported justification for decisions and patient-centered care, emphasizing a whole-person approach and shared decision-making. Further demonstrating a continuum of proficiency, the Framework supported a wider breadth of personal characteristics for some students, including critical analysis, reflection, self-analysis, and adaptability. These personal characteristics illustrate the interwoven cognitive and metacognitive skills that influence and support a high level of clinical reasoning [ 10 , 12 ] and the development of clinical expertise [ 48 , 53 ]. For example [ 54 ], reflection is critical to developing high-level clinical reasoning and advanced level practice [ 12 , 55 ]. Postgraduate students reflected on prior knowledge, experiences, and action within the context of current Framework knowledge, emphasizing active engagement in cognitive processes to inform clinical reasoning processes. Reflection-in-action is highlighted by self-analysis and adaptability. These characteristics require continuous cognitive processing to consider personal strengths and limitations in the context of the patient and evidence-based practice, adapting the clinical encounter as required [ 53 , 55 ]. These findings highlight use of the Framework in postgraduate education to support development of personal characteristics that are indicative of an advanced level of clinical practice [ 12 ].

Synthesis of findings

Derived from synthesis of research study findings and informed by the Postgraduate Musculoskeletal Physiotherapy Practice model [ 12 ], use of the Framework to inform clinical reasoning processes in postgraduate students is illustrated in Fig.  2 . Overlapping clinical reasoning, knowledge and personal characteristic components emphasize the complex interaction of factors contributing to clinical reasoning processes. Personal characteristics of postgraduate students underpin clinical reasoning and knowledge, highlighting their role in facilitating the integration of these two components. Bolded subcomponents indicate convergence of results reflecting all postgraduate students and underscores the variability among postgraduate students contributing to a continuum of clinical reasoning proficiency. The relative weighting of the components is approximately equal to balance the breadth and convergence of subcomponents. Synthesis of findings align with the Postgraduate Musculoskeletal Physiotherapy Practice model [ 12 ], though some differences exist. Limited personal characteristics were identified in this study with little convergence across students, which may be due to the objective of this study and the case analysis approach.

Use of the Framework to inform clinical reasoning in postgraduate level musculoskeletal physiotherapy students. Adapted from the Postgraduate Musculoskeletal Physiotherapy Practice model [ 12 ].

Strengths and limitations

Think aloud case analyses enabled situationally dependent understanding of the Framework to inform clinical reasoning processes in postgraduate level students [ 17 ], considering the rare potential for vascular pathology. A limitation of this approach was the standardized nature of case information provided to students, which may have influenced clinical reasoning processes. Future research studies may consider patient case simulation to address this limitation [ 30 ]. Interviews were conducted during the second half of the postgraduate educational program, and this timing could have influenced clinical reasoning processes compared to if interviews were conducted at the end of the program. Future research can explore use of the Framework to inform clinical reasoning processes in established advanced practice physiotherapists. The sample size of this study aligns with recommendations for think aloud methodology [ 27 , 28 ], achieved rich data, and purposive sampling enabled wide representation of key characteristics (e.g., gender, ethnicity, country of training, physiotherapy experiences), which enhances transferability of findings. Students were aware of the study objective in advance of interviews which may have contributed to a heightened level of awareness of vascular pathology. The prior relationship between students and researchers may have also influenced results, however several strategies were implemented to minimize this influence.

Implications

The Framework is widely implemented within IFOMPT postgraduate educational programs and has led to important shifts in educational curricula [ 9 ]. Findings of this study support use of the Framework as an educational resource in postgraduate physiotherapy programs to inform clinical reasoning processes for safe and effective assessment and management of cervical spine presentations considering the potential for vascular pathology. Individualized approaches may be required to support each student, owing to a continuum of clinical reasoning proficiency. As the Framework was written for practicing musculoskeletal clinicians, future research is required to explore use of the Framework to inform clinical reasoning in learners at different levels, for example entry-level physiotherapy students.

The Framework supported clinical reasoning that used primarily hypothetico-deductive processes in postgraduate physiotherapy students. It informed vascular hypothesis generation in the patient history and testing the vascular hypothesis through patient history questions and selection of physical examination tests, to inform clarity and support for diagnosis and management. Most postgraduate students clinical reasoning processes were characterized as high-level, informed by deep Framework knowledge integrated with a breadth of wider knowledge, and supported by a range of personal characteristics to facilitate the integration of advanced knowledge and high-level clinical reasoning. Future research is required to explore use of the Framework to inform clinical reasoning in learners at different levels.

Data availability

The dataset used and analyzed during the current study are available from the corresponding author on reasonable request.

Safiri S, Kolahi AA, Hoy D, Buchbinder R, Mansournia MA, Bettampadi D et al. Global, regional, and national burden of neck pain in the general population, 1990–2017: systematic analysis of the global burden of Disease Study 2017. BMJ. 2020;368.

Stovner LJ, Nichols E, Steiner TJ, Abd-Allah F, Abdelalim A, Al-Raddadi RM, et al. Global, regional, and national burden of migraine and tension-type headache, 1990–2016: a systematic analysis for the global burden of Disease Study 2016. Lancet Neurol. 2018;17:954–76.

Article   Google Scholar  

Cieza A, Causey K, Kamenov K, Hanson SW, Chatterji S, Vos T. Global estimates of the need for rehabilitation based on the Global Burden of Disease study 2019: a systematic analysis for the global burden of Disease Study 2019. Lancet. 2020;396:2006–17.

Côté P, Yu H, Shearer HM, Randhawa K, Wong JJ, Mior S et al. Non-pharmacological management of persistent headaches associated with neck pain: A clinical practice guideline from the Ontario protocol for traffic injury management (OPTIMa) collaboration. European Journal of Pain (United Kingdom). 2019;23.

Diamanti S, Longoni M, Agostoni EC. Leading symptoms in cerebrovascular diseases: what about headache? Neurological Sciences. 2019.

Debette S, Compter A, Labeyrie MA, Uyttenboogaart M, Metso TM, Majersik JJ, et al. Epidemiology, pathophysiology, diagnosis, and management of intracranial artery dissection. Lancet Neurol. 2015;14:640–54.

Rushton A, Carlesso LC, Flynn T, Hing WA, Rubinstein SM, Vogel S, et al. International Framework for examination of the Cervical Region for potential of vascular pathologies of the Neck Prior to Musculoskeletal intervention: International IFOMPT Cervical Framework. J Orthop Sports Phys Therapy. 2023;53:7–22.

Rushton A, Carlesso LC, Flynn T, Hing WA, Kerry R, Rubinstein SM, et al. International framework for examination of the cervical region for potential of vascular pathologies of the neck prior to orthopaedic manual therapy (OMT) intervention: International IFOMPT Cervical Framework. International IFOMPT Cervical Framework; 2020.

Hutting N, Kranenburg R, Taylor A, Wilbrink W, Kerry R, Mourad F. Implementation of the International IFOMPT Cervical Framework: a survey among educational programmes. Musculoskelet Sci Pract. 2022;62:102619.

Jones MA, Jensen G, Edwards I. Clinical reasoning in physiotherapy. In: Campbell S, Watkins V, editors. Clinical reasoning in the health professions. Third. Philadelphia: Elsevier; 2008. pp. 245–56.

Google Scholar  

Fennelly O, Desmeules F, O’Sullivan C, Heneghan NR, Cunningham C. Advanced musculoskeletal physiotherapy practice: informing education curricula. Musculoskelet Sci Pract. 2020;48:102174.

Rushton A, Lindsay G. Defining the construct of masters level clinical practice in manipulative physiotherapy. Man Ther. 2010;15.

Rushton A, Lindsay G. Defining the construct of masters level clinical practice in healthcare based on the UK experience. Med Teach. 2008;30:e100–7.

Noblet T, Heneghan NR, Hindle J, Rushton A. Accreditation of advanced clinical practice of musculoskeletal physiotherapy in England: a qualitative two-phase study to inform implementation. Physiotherapy (United Kingdom). 2021;113.

Tawiah AK, Stokes E, Wieler M, Desmeules F, Finucane L, Lewis J, et al. Developing an international competency and capability framework for advanced practice physiotherapy: a scoping review with narrative synthesis. Physiotherapy. 2023;122:3–16.

Williams A, Rushton A, Lewis JJ, Phillips C. Evaluation of the clinical effectiveness of a work-based mentoring programme to develop clinical reasoning on patient outcome: a stepped wedge cluster randomised controlled trial. PLoS ONE. 2019;14.

Miles R. Complexity, representation and practice: case study as method and methodology. Issues Educational Res. 2015;25.

Thorne S, Kirkham SR, MacDonald-Emes J. Interpretive description: a noncategorical qualitative alternative for developing nursing knowledge. Res Nurs Health. 1997;20.

Thorne S, Kirkham SR, O’Flynn-Magee K. The Analytic challenge in interpretive description. Int J Qual Methods. 2004;3.

Creswell JW. Research design: qualitative, quantitative, and mixed methods approaches. Sage; 2003.

Dolan S, Nowell L, Moules NJ. Interpretive description in applied mixed methods research: exploring issues of fit, purpose, process, context, and design. Nurs Inq. 2023;30.

Thorne S. Interpretive description. In: Routledge International Handbook of Qualitative Nursing Research. 2013. pp. 295–306.

Thompson Burdine J, Thorne S, Sandhu G. Interpretive description: a flexible qualitative methodology for medical education research. Med Educ. 2021;55.

Tong A, Sainsbury P, Craig J. Consolidated criteria for reporting qualitative research (COREQ): a 32-item checklist for interviews and focus group. Int J Qual Health Care. 2007;19:349–57.

Archibald MM, Ambagtsheer RC, Casey MG, Lawless M. Using zoom videoconferencing for qualitative data Collection: perceptions and experiences of researchers and participants. Int J Qual Methods. 2019;18.

Van Someren M, Barnard YF, Sandberg J. The think aloud method: a practical approach to modelling cognitive. Volume 11. London: Academic; 1994.

Fonteyn ME, Kuipers B, Grobe SJ. A description of think aloud Method and Protocol Analysis. Qual Health Res. 1993;3:430–41.

Lundgrén-Laine H, Salanterä S. Think-Aloud technique and protocol analysis in clinical decision-making research. Qual Health Res. 2010;20:565–75.

Doody C, McAteer M. Clinical reasoning of expert and novice physiotherapists in an outpatient orthopaedic setting. Physiotherapy. 2002;88.

Gilliland S. Physical therapist students’ development of diagnostic reasoning: a longitudinal study. J Phys Therapy Educ. 2017;31.

Ericsson KA, Simon HA. How to study thinking in Everyday Life: contrasting think-aloud protocols with descriptions and explanations of thinking. Mind Cult Act. 1998;5:178–86.

Dwyer SC, Buckle JL. The space between: on being an insider-outsider in qualitative research. Int J Qual Methods. 2009;8.

Shenton AK. Strategies for ensuring trustworthiness in qualitative research projects. Educ Inform. 2004;22:63–75.

Braun V, Clarke V. Using thematic analysis in psychology. Qual Res Psychol. 2006;3:77–101.

Fereday J, Muir-Cochrane E. Demonstrating Rigor using thematic analysis: a Hybrid Approach of Inductive and deductive coding and theme development. Int J Qual Methods. 2006;5.

Elstein ASLSS. Medical problem solving: an analysis of clinical reasoning. Harvard University Press; 1978.

Proudfoot K. Inductive/Deductive Hybrid Thematic Analysis in mixed methods research. J Mix Methods Res. 2023;17.

Charters E. The use of think-aloud methods in qualitative research an introduction to think-aloud methods. Brock Educ J. 2003;12.

Nowell LS, Norris JM, White DE, Moules NJ. Thematic analysis: striving to meet the trustworthiness Criteria. Int J Qual Methods. 2017;16:1–13.

Thurmond VA. The point of triangulation. J Nurs Scholarsh. 2001;33.

Hutting N, Wilbrink W, Taylor A, Kerry R. Identifying vascular pathologies or flow limitations: important aspects in the clinical reasoning process. Musculoskelet Sci Pract. 2021;53:102343.

de Best RF, Coppieters MW, van Trijffel E, Compter A, Uyttenboogaart M, Bot JC, et al. Risk assessment of vascular complications following manual therapy and exercise for the cervical region: diagnostic accuracy of the International Federation of Orthopaedic Manipulative physical therapists framework (the Go4Safe project). J Physiother. 2023;69:260–6.

Petty NJ. Becoming an expert: a masterclass in developing clinical expertise. Int J Osteopath Med. 2015;18:207–18.

Edwards I, Jones M, Carr J, Braunack-Mayer A, Jensen GM. Clinical reasoning strategies in physical therapy. Phys Ther. 2004;84.

Carraccio CL, Benson BJ, Nixon LJ, Derstine PL. Clinical teaching from the Educational Bench to the clinical Bedside: Translating the Dreyfus Developmental Model to the Learning of Clinical Skills.

Benner P. Using the Dreyfus Model of Skill Acquisition to describe and interpret Skill Acquisition and Clinical Judgment in nursing practice and education. Bull Sci Technol Soc. 2004;24:188–99.

Benner P. From novice to expert: Excellence and power in clinical nursing practice. Upper Saddle River, New Jersey: Prentice Hall;: Commemorative Ed; 2001.

Jensen GM, Gwyer J, Shepard KF, Hack LM. Expert practice in physical therapy. Phys Ther. 2000;80.

Huhn K, Gilliland SJ, Black LL, Wainwright SF, Christensen N. Clinical reasoning in physical therapy: a Concept Analysis. Phys Ther. 2019;99.

Hutting N, Kranenburg HA, Rik KR. Yes, we should abandon pre-treatment positional testing of the cervical spine. Musculoskelet Sci Pract. 2020;49:102181.

Kranenburg HA, Tyer R, Schmitt M, Luijckx GJ, Schans C, Van Der, Hutting N, et al. Effects of head and neck positions on blood flow in the vertebral, internal carotid, and intracranial arteries: a systematic review. J Orthop Sports Phys Ther. 2019;49:688–97.

Hutting N, Kerry R, Coppieters MW, Scholten-Peeters GGM. Considerations to improve the safety of cervical spine manual therapy. Musculoskelet Sci Pract. 2018;33.

Wainwright SF, Shepard KF, Harman LB, Stephens J. Novice and experienced physical therapist clinicians: a comparison of how reflection is used to inform the clinical decision-making process. Phys Ther. 2010;90:75–88.

Dy SM, Purnell TS. Key concepts relevant to quality of complex and shared decision-making in health care: a literature review. Soc Sci Med. 2012;74:582–7.

Christensen N, Jones MA, Higgs J, Edwards I. Dimensions of clinical reasoning capability. In: Campbell S, Watkins V, editors. Clinical reasoning in the health professions. 3rd edition. Philadelphia: Elsevier; 2008. pp. 101–10.

Download references

Acknowledgements

The authors would like to acknowledge study participants and the transcriptionist for their time in completing and transcribing think aloud interviews.

No funding was received to conduct this research study.

Author information

Authors and affiliations.

School of Physical Therapy, Western University, London, Ontario, Canada

Katie L. Kowalski, Heather Gillis, Katherine Henning, Paul Parikh, Jackie Sadi & Alison Rushton

You can also search for this author in PubMed   Google Scholar

Contributions

Katie Kowalski: Conceptualization, methodology, validation, formal analysis, investigation, data curation, writing– original draft, visualization, project administration. Heather Gillis: Validation, resources, writing– review & editing. Katherine Henning: Investigation, formal analysis, writing– review & editing. Paul Parikh: Validation, resources, writing– review & editing. Jackie Sadi: Validation, resources, writing– review & editing. Alison Rushton: Conceptualization, methodology, validation, writing– review & editing, supervision.

Corresponding author

Correspondence to Katie L. Kowalski .

Ethics declarations

Ethics approval and consent to participate.

Western University Health Science Research Ethics Board granted ethical approval (Project ID: 119934). Participants provided written informed consent prior to participating in think aloud interviews.

Consent for publication

Not applicable.

Competing interests

Author AR is an author of the IFOMPT Cervical Framework. Authors JS and HG are lecturers on the AHCP CMP program. AR and JS led standardized teaching of the Framework. Measures to reduce the influence of potential competing interests on the conduct and results of this study included: the Framework representing international and multiprofessional consensus, recruitment of participants by author KK, data collection and analysis completed by KK with input from AR, JS and HG at the stage of data synthesis and interpretation, and wider peer scrutiny of initial findings. KK, KH and PP have no potential competing interests.

Authors’ information

The lead author of this study (AR) is the first author of the International IFOMPT Cervical Framework.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Supplementary material 2, supplementary material 3, supplementary material 4, rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

Kowalski, K.L., Gillis, H., Henning, K. et al. Use of the International IFOMPT Cervical Framework to inform clinical reasoning in postgraduate level physiotherapy students: a qualitative study using think aloud methodology. BMC Med Educ 24 , 486 (2024). https://doi.org/10.1186/s12909-024-05399-x

Download citation

Received : 11 February 2024

Accepted : 08 April 2024

Published : 02 May 2024

DOI : https://doi.org/10.1186/s12909-024-05399-x

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• International IFOMPT Cervical Framework
• Clinical reasoning
• Postgraduate students
• Physiotherapy
• Educational research
• Qualitative research
• Think aloud methodology

BMC Medical Education

ISSN: 1472-6920

• Submission enquiries: [email protected]
• General enquiries: [email protected]

Creativity and the Brain: How to Be a Creative Thinker

What do we know from research on brain activity involved in creative thought.

Posted April 30, 2024 | Reviewed by Michelle Quirk

• The book "The Creative Act" argues that creativity is a skill we can all use daily.
• Creativity is complex and involves multiple brain regions.
• Research shows that there are several ways to improve our creative thinking.

This post is part 2 of a series.

In my previous post, I wrote that, after being inspired by Rick Rubin’s book, The Creative Act: A Way of Being, I decided to find out what is going on in the human brain that results in creativity. It turned out to be a very complex and complicated subject. That is mainly because it is difficult to clearly define creativity, and there are many different kinds of creative processes, such as visual art, music, creative thinking , etc.

Coming from the field of cognitive processes, I decided that I would concentrate on research related to brain activity involved in creative thought processes. Most of the time, cognitive creativity involves testing the person’s divergent thinking (generating possible solutions to the problem) or convergent thinking (finding a single, correct solution to the problem).

The review of research papers indicated that creative thinking (convergent and divergent thinking) requires the coordination of multiple brain regions, mainly the executive control network (simply speaking involves planning, organizing, problem-solving, and decision-making ), default mode network (areas of the brain that are activated when we are letting our minds wander at rest), and salience network (a network that is involved in the awareness of the feelings associated with rewards). But, obviously, other parts of the brain are also involved, and this depends on the specific goal/outcome that we want to achieve.

I also promised my readers that I would try to find answers to the question of how to be a creative thinker. There are many suggestions on the internet, but let’s see what the research says.

You can learn how to meditate and practice it daily.

It may come as a surprise to many people, but the majority of the research papers in that area point to the daily practice of meditation as a way to improve creative thinking. It is not a surprise to me because I am a believer in meditation and do it daily. I also encourage all my patients to try to do it daily.

In a Chinese study (Ding, X. et al. 2014), 40 Chinese undergraduate students were assigned to three groups, a meditation group (30 minutes daily for 7 days), a relaxation training group, and a control group. Creativity performance was assessed by the Torrance Test of Creative Thinking (TTCT). The results indicated that the subjects in the meditation group improved their creativity performance on the divergent thinking tasks.

Research studies on meditation also indicate that it helps improve attention/ concentration skills and emotional regulation and reduces stress and anxiety , so it looks like a good daily habit to start.

You can read aloud and do arithmetic calculations.

In a Taiwan study (Lin, WL. et al. 2018), 50 junior high students were divided into a training group or a control group. The training group was reading aloud and performing arithmetic calculations for 20 sessions. The control group played the game Tetris (a puzzle video game). The results indicated that the participants in the training group outperformed the control group in thinking and creative abilities.

You can do neurofeedback.

Neurofeedback is a computer-guided, noninvasive brain-function training based on electroencephalography (EEG) feedback. Neurofeedback is also called neurotherapy, neurobiofeedback, or EEG biofeedback, and it helps control involuntary processes such as muscle tension and heart rate. Usually, the person is responding to a computer display of her/his own electrical activity of the brain, but it may also simply be a sound stimulation. The most important factor is that neurofeedback focuses on helping a person train himself/herself to regulate brain functions.

In an Italian study (Agnoli, S. et al. 2018), 80 female students from the University of Bologna got three neurofeedback training sessions. The researchers also measured the participants’ lifetime creative achievement by using the Creative Activity and Accomplishment Checklist. The results were measured with the divergent thinking tasks (producing original and effective ideas). The results indicated an increase in both originality and fluency. The increase was particularly evident in participants with an initial low creative achievement level.

This is good news for people who believe that they are not that creative. You may get better with neurofeedback training sessions. Artists and athletes do this nowadays to enhance their performance.

You can do overinclusive thinking training.

Overinclusive thinking can be described as increased generalization and/or considering concepts that most people consider unrelated to certain categories, which provides an increased number of options. In a Taiwan study (Chiu, F.C. 2015), the researcher examined the effect of overinclusive thinking on creativity. Four experiments were designed, and the subjects were undergraduate students who were randomly assigned to an overinclusive thinking training group or a control group. The training group did better on the overinclusive thinking that is related to creativity. The fluency and originality performance were higher than in the control group and the insight problem-solving was also better than in the control group.

So, if you would like to be a creative thinker, you can try some of the ideas described above. Good luck on the road to creativity!

Rick Rubin. The Creative Act: A Way of Being . Penguin Press, NY 2023.

Ding, X. et al. “Improving creativity performance by short-term meditation” Behavioral and Brain Functions. Vol. 10, 2014.

Lin, WL. et al. “ Improving junior high students’ thinking and creative abilities with an executive function training program” Thinking Skills and Creativity . Vol. 29, Sept. 2018.

Agnoli, S. et al. “Enhancing creative cognition with a rapid right-parietal neurofeedback procedure.” Neuropsychologia, Vol. 118, Part A Sept. 2018.

Chiu, F.C. “ Improving your creative potential without awareness: Overinclusive thinking training.” Thinking Skills and Creativity . Vol 15. March 2015.

Barbara Koltuska-Haskin, Ph.D., is a neuropsychologist in Albuquerque, New Mexico and the author of How My Brain Works.

• Find Counselling
• Find a Support Group
• Find Online Therapy
• United Kingdom
• Asperger's
• Bipolar Disorder
• Chronic Pain
• Eating Disorders
• Passive Aggression
• Personality
• Goal Setting
• Positive Psychology
• Stopping Smoking
• Low Sexual Desire
• Relationships
• Child Development
• Therapy Center NEW
• Diagnosis Dictionary
• Types of Therapy

Understanding what emotional intelligence looks like and the steps needed to improve it could light a path to a more emotionally adept world.

• Emotional Intelligence
• Gaslighting
• Affective Forecasting
• Neuroscience

We’re All Reading Wrong

To access the full benefits of literature, you have to share it out loud.

Listen to this article

Produced by ElevenLabs and News Over Audio (NOA) using AI narration.

Updated at 4:32 p.m. ET on May 3, 2024

Reading, while not technically medicine, is a fundamentally wholesome activity. It can prevent cognitive decline , improve sleep , and lower blood pressure . In one study, book readers outlived their nonreading peers by nearly two years. People have intuitively understood reading’s benefits for thousands of years: The earliest known library , in ancient Egypt, bore an inscription that read The house of healing for the soul .

But the ancients read differently than we do today. Until approximately the tenth century , when the practice of silent reading expanded thanks to the invention of punctuation, reading was synonymous with reading aloud. Silent reading was terribly strange, and, frankly, missed the point of sharing words to entertain, educate, and bond. Even in the 20th century, before radio and TV and smartphones and streaming entered American living rooms, couples once approached the evening hours by reading aloud to each other.

But what those earlier readers didn’t yet know was that all of that verbal reading offered additional benefits: It can boost the reader’s mood and ability to recall . It can lower parents’ stress and increase their warmth and sensitivity toward their children. To reap the full benefits of reading, we should be doing it out loud, all the time, with everyone we know.

Reading aloud is a distinctive cognitive process, more complex than simply reading silently, speaking, or listening. Noah Forrin, who researched memory and reading at the University of Waterloo, in Canada, told me that it involves several operations—motor control, hearing, and self-reference (the fact that you said it)—all of which activate the hippocampus, a brain region associated with episodic memory. Compared with reading silently, the hippocampus is more active while reading aloud, which might help explain why the latter is such an effective memory tool. In a small 2012 study , students who studied a word list remembered 90 percent of the words they’d read aloud immediately afterward, compared with 71 percent of those they’d read silently. (One week later, participants remembered 59 percent of the spoken words and 48 percent of the words read silently.)

So although you might enjoy an audiobook narrated by Meryl Streep, you would remember it better if you read parts of it out loud—especially if you did so in small chunks, just a short passage at a time, Forrin said. The same goes for a few lines of a presentation that you really want to nail. Those memory benefits hold true whether or not anyone is around to hear your performance.

Verbal reading without an audience is, in fact, surprisingly common. While studying how modern British people read aloud, Sam Duncan, a professor of adult literacies at University College London, found that they read aloud—and alone—for a variety of reasons. One woman recited Welsh poetry to remember her mother, with whom she spoke Welsh as a girl. One young man read the Quran out loud before work to better understand its meaning. Repeating words aloud isn’t just key to memorization, Duncan told me—it can be key to identity formation too.

From the August 1904 issue: On reading aloud

Plenty of solitary vocal reading no doubt consists of deciphering recipes and proofreading work emails, but if you want to reap the full perks, the best selections are poetry and literature. These genres provide access to facets of human experience that can be otherwise unreachable, which helps us process our own emotions and memories, says Philip Davis, an emeritus professor of literature and psychology at the University of Liverpool. Poetry, for example, can induce peak emotional responses , a strong reaction that might include goose bumps or chills. It can help you locate an emotion within yourself, which is important to health as a form of emotional processing.

Poetry also contains complex, unexpected elements, like when Shakespeare uses god as a verb in Coriolanus : “This last old man … godded me.” In an fMRI study that Davis co-authored in 2015, such literary surprise was shown to be stimulating to the brain. Davis told me that literature, with its “mixture of memory and imagination,” can cause us to recall our most complex experiences and derive meaning from them. A poem or story read aloud is particularly enthralling, he said, because it becomes a live presence in the room, with a more direct and penetrative quality, akin to live music. Davis likens the role of literature and live reading to a spark or renewal, “a bringing of things back to life.”

Discussing the literature that you read aloud can be particularly valuable. Davis told me doing so helps penetrate rigid thinking and can dislodge dysfunctional thought patterns. A qualitative 2017 study led by Josie Billington at the University of Liverpool found that, for those who have chronic pain and the depression that tends to come with it, such discussion expands emotional vocabulary —a key tenet of psychological well-being— perhaps even more so than cognitive behavioral therapy . (The allure of an audience has one notable exception: If you’re anxious, reading aloud can actually reduce memory and comprehension . To understand this effect, one need only harken back to fifth grade when it was your turn to read a paragraph on Mesopotamia in class.)

Read: How to keep your book club from becoming a wine club

The health benefits of reading aloud are so profound that some doctors in England now refer their chronic-pain patients to read-aloud groups. Helen Cook, a 45-year-old former teacher in England, joined one of these groups in 2013. Cook had a pelvic tumor that had sent anguish ricocheting through her hip and back for a decade, and medication never seemed to help. Before she joined the reading group, Cook had trouble sleeping, lost her job, and “had completely lost myself,” she told me. Then, she and nine other adults began working their way through some 300 pages of Hard Times , by Charles Dickens.

Cook told me she recognized her experience in the characters’ travails, and within months, she “rediscovered a love for life,” even returning to college for a master’s degree in literature. She’s not the only one who found relief: In Billington’s 2017 study, everyone who read aloud in a group felt emotionally better and reported less pain for two days afterward.

Hearing words read aloud to you also has unique advantages, especially for kids. Storytelling has been shown to increase hospitalized children’s levels of oxytocin while decreasing cortisol and pain. Julie Hunter, who for more than 20 years has taught preschool kids (including my daughter), told me that interactive reading increases young children’s comprehension , builds trust , and enhances social-emotional skills . A recent study by researchers at the Brookings Institution found that children smiled and laughed more when being read to by a parent than when listening to an automatically narrated book alone.

Read: An ode to being read to

Anecdotal evidence suggests that adults, too, can benefit from such listening. For 25 years, Hedrick and Susan Smith, ages 90 and 84, respectively, have read more than 170 books aloud. They started by reading in the car, to pass the time, but it was so much fun that they started reading every night before they turned out the light, Hedrick told me. Together, they tried to comprehend One Hundred Years of Solitude , narrated Angela’s Ashes in four different Irish accents, and deduced clues in John le Carré thrillers. They felt more connected, and went to sleep in brighter moods. If they liked the book, they couldn’t wait for the other to read the next chapter aloud—even, and perhaps especially, when the sound of the other’s voice sent them off to sleep.

Due to an editing error, this article originally misidentified the author of a 2017 study.