Research Article: Social intuition as a form of implicit learning: Sequences of body movements are learned less explicitly than letter sequences

Date Published: May 21, 2012

Publisher: University of Finance and Management in Warsaw

Author(s): Elisabeth Norman, Mark C. Price.


In the current paper, we first evaluate the suitability of traditional serial
reaction time (SRT) and artificial grammar learning (AGL) experiments for
measuring implicit learning of social signals. We then report the results of a
novel sequence learning task which combines aspects of the SRT and AGL paradigms
to meet our suggested criteria for how implicit learning experiments can be
adapted to increase their relevance to situations of social intuition. The
sequences followed standard finite-state grammars. Sequence learning and
consciousness of acquired knowledge were compared between 2 groups of 24
participants viewing either sequences of individually presented letters or
sequences of body-posture pictures, which were described as series of yoga
movements. Participants in both conditions showed above-chance classification
accuracy, indicating that sequence learning had occurred in both stimulus
conditions. This shows that sequence learning can still be found when learning
procedures reflect the characteristics of social intuition. Rule awareness was
measured using trial-by-trial evaluation of decision strategy (Dienes & Scott, 2005; Scott & Dienes, 2008). For letters,
sequence classification was best on trials where participants reported
responding on the basis of explicit rules or memory, indicating some explicit
learning in this condition. For body-posture, classification was not above
chance on these types of trial, but instead showed a trend to be best on those
trials where participants reported that their responses were based on intuition,
familiarity, or random choice, suggesting that learning was more implicit.
Results therefore indicate that the use of traditional stimuli in research on
sequence learning might underestimate the extent to which learning is implicit
in domains such as social learning, contributing to ongoing debate about
levels of conscious awareness in implicit learning.

Partial Text

Implicit learning is assumed to play a central role in various everyday behaviours.
One example is the learning of complex patterns of motor responses involved in
skills like playing musical instruments and driving, in which the details of the
acquired knowledge are not fully accessible to conscious awareness (Clegg, DiGirolamo, & Keele, 1998). Another
example is the acquisition of grammatical rules of one’s native language,
which is claimed to occur largely independently of the conscious intent of the
learner (Cleeremans, Destrebecqz, & Boyer,
1998; Reber, 1967, 1989). Yet another category of everyday
behaviours explained in terms of implicit learning is the encoding and decoding of
social signals in social interactions (Lieberman,
2000). For example, when people are sometimes able to accurately judge
the personality of another person without being able to verbalize what the judgement
was based on, this may be explained in terms of complex behavioural regularities
being learned without full conscious awareness (Lewicki,Hill, & Czyzewska, 1992). Because of its assumed central
role in social cognition, implicit learning has even been referred to as the
cognitive substrate of social intuition (Lieberman,
2000). According to Lieberman, social intuition involves making rapid
judgements about the emotions, personality, intentions, attitudes, and skills of
others (p. 111). Such judgements are often based on the perception of sequences of
various forms of nonverbal cues, including subtle facial expressions, body postures,
and nonverbal gestures. Lieberman refers to this process as the “learning of
nonverbal decoding.”This is regarded as an acquired ability that develops
continuously throughout the life span.

In response to the methodological criteria outlined above, we conducted an experiment
which combines aspects of both the SRT and the AGL paradigms in a novel manner. The
experiment measured implicit learning of sequences of pictures of body postures that
were presented individually but followed an artificial grammar structure. As in
traditional SRT tasks, the sequences could be taken to represent a dynamic
transformation of the state of a single entity (Suggestions 1 and 2). But as in
traditional AGL tasks, participants were exposed to a series of discrete exemplars
of the sequence regularity, and sequences were not completely identical to each
other (Suggestions 3 and 4). The degree of learning was operationalised in terms of
classification accuracy during a subsequent test phase where a series of novel
sequences were presented. To distinguish between nonconscious, intuitive and fully
explicit learning, the study also included detailed measurement of the degree of
conscious awareness of acquired knowledge as commonly applied within SRT and AGL
tasks (Suggestion 5).

Mean classification accuracy, as measured by the proportion of trials on which a
correct response was made, was significantly above the chance level of .5 both in
the yoga condition, t(21) = 2.50, p = .02
(two-tailed), and in the letter condition, t(23) = 4.76,
p < .001 (two-tailed). A 2 × 2 ANOVA compared classification accuracy between these two conditions and between participants trained on Grammars A or B. Since training on Grammar A versus B had no significant main effect (p = .45)and did not interact with the yoga/letter manipulation (p = .36), this variable was excluded from subsequent analyses. However, mean classification accuracy was significantly higher in the letter condition (M = 0.59, SE = 0.02) than in the yoga condition, M = 0.53, SE = 0.01, F(1, 42) = 6.39, p = .02. The amount of learning and the degree of consciousness over what was learned was explored in each of two experimental conditions (yoga sequences vs. letter sequences) in an implicit learning experiment that combined procedural aspects of both the SRT and AGL tasks. During training, participants viewed sequences of stimuli that followed a rule based structure. Several aspects of the general procedure and stimulus design were intended to improve the suitability of the implicit learning experiment as an analog of learning involved in social intuition, namely: (a) the dynamic nature of the sequences, in which sequence elements appeared one by one; (b) the presentation of each sequence as a discrete exemplar; (c) the use of varying sequences that adhered to an artificial grammar rule rather than using cyclic repetition of an identical sequence. In addition, our yoga stimulus condition, which consisted of sequences of body postures, used more socially naturalistic stimuli than the letter sequences typically used in AGL tasks or the abstract shapes typically used in SRT tasks, and also depicted the transformation of a single entity (a body) rather than a juxtaposition of separate letters.   Source: