Thin Slicing

Can We Really Judge Behaviour in Just a Few Minutes?

Have you ever made a snap judgement about someone within the first minute of meeting them—and later found it to be remarkably accurate? That might be an example of thin slicing in action.

What Is Thin Slicing?

The term thin slicing was popularised by psychologists Nalini Ambady and Robert Rosenthal in a 1992 meta-analysis. They demonstrated that observers could make surprisingly accurate judgements about a person’s personality, emotions, and interpersonal dynamics based on brief snippets of behaviour. These thin slices—sometimes as short as 30 seconds—were found to predict outcomes such as deception detection, teacher effectiveness, trustworthiness, and even relationship satisfaction.

Ambady and Rosenthal’s work sparked a wealth of research. One notable study by Murphy (2007) showed that just one minute of observation was enough for people to make accurate assessments of a person’s intelligence. Similarly, Roter et al. (2011) found that one-minute slices reliably reflected communication styles in medical consultations lasting over ten minutes. Hirschmann et al. (2018) went further, comparing 10-minute and 40-minute samples, and concluded that both were equally valid for assessing maternal sensitivity.

Why Use Thin Slices?

Behavioural coding—systematically analysing observed behaviours—is time-intensive and labour-intensive. Researchers often make real-time or frame-by-frame notes using detailed coding schemes. This can become even more demanding when dealing with complex behavioural categories or introducing new coding systems (Pesch & Lumeng, 2017). Given the considerable resources required, it’s no surprise that researchers often opt to code only short portions of longer interactions. As James et al. (2012) noted, this has become common practice in behavioural research.

Thin slicing offers a practical solution: a way to reduce workload without necessarily sacrificing accuracy. It also aligns with how people naturally make snap judgements.

The Predictive Power—and Limits—of Thin Slices

There is a growing body of evidence supporting the idea that thin slices can predict broader outcomes. Some of the more famous examples include:

• First Impressions: Ambady et al. (1999) found that students could accurately rate the effectiveness of teachers based on silent, 10-second video clips.

• Relationship Dynamics: Gottman and Levenson (1992) famously predicted divorce with over 90% accuracy using brief samples of couple interactions, focusing on micro-behaviours like contempt and defensiveness.

• Deception Detection: Vrij (2008) showed that people can sometimes detect deception from short segments of nonverbal behaviour—especially when focusing on emotional leakage rather than verbal content.

However, we must guard against confirmation bias, which may account for some perceived consistency. If our initial impression of a teacher is negative, we may interpret any classroom behavioural issues as evidence of poor teaching. Conversely, if we initially judge the teacher favourably, we might attribute the same issues to the students instead.

Murphy (2005) found moderate to high correlations between 1- to 3-minute slices and full 15-minute interactions of nonverbal behaviour. Similarly, Carcone et al. (2015) showed that the proportion of verbal behaviours in 1- to 2-minute slices closely matched those in full 30-minute sessions. However, not all findings are consistent. Some studies caution against overgeneralising results from thin slice data. For instance, James et al. (2012) found that contingency measures of vocal and visual behaviours gathered from 3- or 6-minute slices differed significantly from those derived from full 18-minute sessions. Murphy et al. (2019) and Wang et al. (2020) also observed reduced predictive validity in certain contexts.

Real-World Applications

The implications of thin slicing are far-reaching. In education, thin slices are used to assess teacher performance with minimal observation time. In healthcare, patient–provider interactions are increasingly evaluated using short video clips to ensure quality communication. Recruiters, therapists, and even intelligence agencies often rely on brief interactions to make crucial decisions.

In the realm of body language and nonverbal communication, thin slicing reinforces the idea that we can’t not communicate. Everything from posture shifts to facial touching can be revealing in just a few seconds. It seems that our initial gut instincts can serve us well.

But Proceed with Caution

Despite its promise, thin slicing is not a magic bullet. It should be applied thoughtfully. Its reliability depends on the behaviour being measured, the observer’s expertise, and the context of the interaction. Like any research method, thin slicing has limitations, and further empirical guidance is needed to establish best practices across different domains.

Conclusion

In many cases, the longer the slice, the more accurate the representation. However, thin slicing challenges the assumption that more data always leads to better understanding. Sometimes, a minute is all we need. Yet, like all tools, its effectiveness depends on how—and where—it is applied. For those of us interested in body language, first impressions, and the hidden patterns of human interaction, thin slicing opens up a fascinating window into how much we communicate in the blink of an eye.

References

Ambady, N., & Rosenthal, R. (1992). Thin slices of expressive behaviour as predictors of interpersonal consequences: A meta-analysis. Psychological Bulletin, 111(2), 256–274.

Beebe, B., Jaffe, J., Markese, S., Buck, K., Chen, H., Cohen, P., ... & Andrews, H. (2011). The origins of 12-month attachment: A microanalysis of 4-month mother–infant interaction. Attachment & Human Development, 13(1), 1–27.

Carcone, A. I., Tokarz, R., & Naar, S. (2015). Short and sweet: Concordance between thin slice and full-length motivational interviewing session coding. Journal of Substance Abuse Treatment, 55, 44–50.

Feldman, R., & Eidelman, A. I. (2007). Maternal postpartum behaviour and the emergence of infant–mother and infant–father synchrony in preterm and full-term infants: The role of neonatal vagal tone. Developmental Psychobiology, 49(3), 290–302.

Gottman, J. M., & Levenson, R. W. (1992). Marital processes predictive of later dissolution: Behaviour, physiology, and health. Journal of Personality and Social Psychology, 63(2), 221–233.

Hirschmann, J., Nijssens, L., van Bakel, H. J., & Cillessen, A. H. (2018). Observing maternal sensitivity: Comparing 10-minute and 40-minute observations. Infant Mental Health Journal, 39(6), 663–677.

James, K., Surkin, L., & Sirois, F. M. (2012). Sampling interaction behaviour: A comparison of thin slices and full-length interactions. Canadian Journal of Behavioural Science, 44(4), 309–318.

Murphy, N. A. (2005). Using thin slices for behavioural coding. Journal of Nonverbal Behavior, 29(3), 235–246.

Murphy, N. A. (2007). Person perception from thin slices of behaviour: A replication and extension. Journal of Nonverbal Behavior, 31(2), 73–84.

Murphy, N. A., Hall, J. A., & Colvin, C. R. (2019). Accuracy of judging personality traits from thin slices of behaviour: A meta-analytic review. Personality and Social Psychology Bulletin, 45(5), 674–688.

Northrup, J. B., & Iverson, M. L. (2020). Emotional expression in brief interactions: Predictive validity of thin slices. Emotion, 20(6), 989–997.

Pesch, M. H., & Lumeng, J. C. (2017). Measuring parent–child interactions: Comparing thin slice and full-session coding. Early Child Development and Care, 187(7), 1160–1170.

Roter, D. L., Larson, S., & Beach, M. C. (2011). Thin slice ratings of patient-centred communication: A valid tool for use in patient care. Patient Education and Counseling, 82(3), 448–454.

Vrij, A. (2008). Detecting Lies and Deceit: Pitfalls and Opportunities. Wiley.

Wang, L., Jiang, W., & Shen, L. (2020). Judging emotions from thin slices: Do duration and context matter? Journal of Nonverbal Behavior, 44(2), 179–195.