Are Thin Slices Of Nonverbal Behaviour Enough?
Christopher Philip
A team of researchers led by Nora Murphy, Loyola Marymount University, Los Angeles, devised a study to test if observing small amounts of behaviour occurring over a short period of time was as accurate as reading larger amounts over an extended period of time.
Earlier research has found that faces exposed for mere microseconds can accurately produce readings of extraversion and even sexual orientation. However, the researchers wondered if very brief views of behaviour could lead to similarly accurate predictions.
Toward this end, the researchers set up four studies to examine five main ideas including:
(a) How well individual slices of a given behavior predict other slices in the same interaction;
(b) How well a slice of a given behavior represents the entirety of that behavior within an interaction;
(c) How long a slice is necessary to sufficiently represent the entirety of a behavior within an interaction;
(d) Which slices best capture the entirety of behavior, across different behaviors;
(e) Which behaviors (of six measured behaviors) are best captured by slices.
The context of the four studies included:
(1) A laboratory dyatic interactions – a brief “get-to-know-you session”
(2) A mock job interview
(3) A real (not mock) job interview
(4) A mock medical/patient interaction.
The results found that 1.5 minute slices of behaviour from the beginning of an interaction was enough to make general inferences about certain behaviours. The cues that were most reliable include gaze and nods. “These results,” say the researchers “indicate that individual thin slices of these behaviors predicted other individual slices of the same behavior reasonably well on average.”
The researchers also tracked gestures, self-touch and smile and note that while these were not as reliable throughout the situations, they were still relative consistent from one slice to another. The final observation related to speaking time. This was the poorest performer and produced the least reliable results between slices.
So how long should you observe someone to get a sense of their use of each particular behaviour?
Gaze was the strongest across the studies. The results suggest that observations of 30 seconds to 1 minute is enough to make reliable inferences about this behaviour. Nods and smiles were next but it was not as strong as gaze. Gestures, self-touch and speaking time was much more variable. Thus, observations of 30 seconds to 1 minute would not be enough to make reliable judgments.
The data clearly show that it is not necessary to code the entire duration of behavior to have equivalent cumulative validity,” say the researchers “the first 1.30-2.00 min of interactions appear to adequately represent the behaviors of gaze, gestures, nods, self-touch, and smiles.”
Speaking time required the most amount of time to produce validity at 3.5-4 minutes.
Overall the study helps one decide just how long is needed to make a reliable judgments of another person without having to sacrifice time in doing so. This is particularly applicable to managers looking to hire a suitable candidate whom is using nonverbal reading to aid in their decisions. As shown here, gaze and nodding requires only a short period of observation to get a sense of how a person uses these generally. Other attributes including self-touch and smiles were not as reliable as gaze and nodding, but adequate. Speaking time had the worst thin slice reliability and therefore required the most amount of time to be properly assessed. Overall, the study found that 1.5-2 minutes was enough to make fairly accurate predictions about the given behaviours measured herein.
Resources
Murphy, Nora A.; Judith A. Hall; Marianne Schmid Mast; Mollie A. Ruben; Denise Frauendorfer; Danielle Blanch-Hartigan; Debra L. Roter and Laurent Nguyen. Reliability and Validity of Nonverbal Thin Slices in Social Interactions. Personality and Social Psychology Bulletin. 2019 41(2): 199-213.