TY - JOUR

T1 - Social feedback interferes with implicit rule learning: Evidence from event-related brain potentials

AU - Beston, Philippa

AU - Barbet, Cécile

AU - Heerey, Erin

AU - Thierry, Guillaume

N1 - This work was supported by the Swiss National Science Foundation [grant numbers P2NEP1_155426, P300P1_164558 to C.B.]

PY - 2018/12/18

Y1 - 2018/12/18

N2 - The human brain can learn contingencies built into stimulus sequences unconsciously. The quality of such implicit learning has been connected to stimulus social relevance, but results so far are inconsistent. We engaged participants in an implicit-intentional learning task in which they learned to discriminate between legal and illegal card triads on the sole basis of feedback provided within a staircase procedure. Half of the participants received feedback from pictures of faces with a happy or sad expression (social group) and the other half based on traffic light icons (symbolic group). We hypothesised that feedback from faces would have a greater impact on learning than that from traffic lights. Although performance during learning did not differ between groups, the feedback error-related negativity (fERN) was delayed by ~20 ms for social relative to symbolic feedback, and the P3b modulation elicited by infrequent legal card triads within a stream of illegal ones during the test phase was significantly larger in the symbolic than the social feedback group. Furthermore, the P3b mean amplitude recorded at test negatively correlated with the latency of the fERN recorded during learning. These results counterintuitively suggest that, relative to symbolic feedback, socially salient feedback interferes with implicit learning.

AB - The human brain can learn contingencies built into stimulus sequences unconsciously. The quality of such implicit learning has been connected to stimulus social relevance, but results so far are inconsistent. We engaged participants in an implicit-intentional learning task in which they learned to discriminate between legal and illegal card triads on the sole basis of feedback provided within a staircase procedure. Half of the participants received feedback from pictures of faces with a happy or sad expression (social group) and the other half based on traffic light icons (symbolic group). We hypothesised that feedback from faces would have a greater impact on learning than that from traffic lights. Although performance during learning did not differ between groups, the feedback error-related negativity (fERN) was delayed by ~20 ms for social relative to symbolic feedback, and the P3b modulation elicited by infrequent legal card triads within a stream of illegal ones during the test phase was significantly larger in the symbolic than the social feedback group. Furthermore, the P3b mean amplitude recorded at test negatively correlated with the latency of the fERN recorded during learning. These results counterintuitively suggest that, relative to symbolic feedback, socially salient feedback interferes with implicit learning.

KW - Event-related potentials

KW - Implicit learning

KW - Social Feedback

KW - P3b

KW - fERN

U2 - 10.3758/s13415-018-0635-z

DO - 10.3758/s13415-018-0635-z

M3 - Article

VL - 18

SP - 1248

EP - 1258

JO - Cognitive, Affective, and Behavioral Neuroscience

JF - Cognitive, Affective, and Behavioral Neuroscience

SN - 1530-7026

IS - 6

ER -