Metadata language
Instytut Biologii Doświadczalnej im. Marcelego Nenckiego PAN
Contributor:Szatkowska, Iwona (1961- ) : Supervisor
Publisher:Instytut Biologii Doświadczalnej im. Marcelego Nenckiego PAN
Place of publishing: Date issued/created: Description:110 pages : illustrations ; 30 cm ; Bibliography ; Summaries in English
Degree name: Degree discipline : Degree grantor:Nencki Institute of Experimental Biology PAS ; degree obtained: 10.10.2025
Type of object: Subject and Keywords:Associative learning ; Basolateral amygdala ; Centromedial amygdala ; fMRI ; Humans prediction error
Abstract:
Associative learning involves forming associations between a neutral stimulus and a reinforcer (as in classical conditioning) or between a behavior and the outcome it produces (as in instrumental conditioning). Modern theories suggest that associative learning arises from a prediction error (PE) defined as a discrepancy between expected and actual reinforcement. A positive PE occurs when the actual reinforcement exceeds expectations, strengthening the association, whereas a negative PE happens when the reinforcement is less than anticipated, weakening the association. These prediction errors can further be categorized as either appetitive, relating to rewards, or aversive, associated with punishments. The amygdala is critical for associative learning, particularly in signaling prediction errors. Research on animals indicates that different regions of the amygdala, like the centromedial amygdala (CMA) and basolateral amygdala (BLA), serve distinct functions in associative learning. However, the precise functional roles of these areas in humans remain largely unexplored. Therefore, this study aimed to investigate the specific contributions of the CMA and BLA in the left and right hemispheres of the human brain to the process of associative learning. The fMRI study included both classical learning (experiment I) and instrumental learning (experiment II) in the appetitive and aversive contexts. In both experiments compound reinforcers comprised a gustatory (sweet, salty or tasteless liquid) and a social component (a 3-second video of a person drinking a pleasant, unpleasant, or neutral beverage). In experiment I, the participants (N = 37, 20 females) were tasked with predicting the type of reinforcement based on the cue presented on the screen. In experiment II, the participants (N = 33, 16 females) were asked to independently choose one of two simultaneously presented cues. Their responses were used to compute prediction error values according to the Rescorla- Wagner learning model, and these prediction error values were then applied as a parametric modulator of the BOLD signal. The findings revealed that the CMA in the left hemisphere is involved in signaling negative prediction errors during both appetitive and aversive classical learning, as well as negative and positive prediction errors during appetitive instrumental learning. Furthermore, the CMA in the left hemisphere was that amygdala region whose activity correlated with personality traits, such as extraversion and neuroticism, and with BMI. In the right hemisphere, CMA activity was observed in relation to negative prediction errors during appetitive classical learning. Additionally, the BLA in the left hemisphere showed activity specifically linked to positive prediction errors during appetitive instrumental learning. The findings suggest that the CMA and BLA regions in the left and right hemispheres of the brain are engaged differently in associative learning tasks involving gustatory-social reinforcements. The CMA in the left hemisphere appears to play a pivotal and universal role, likely related to adjusting the association strength between cues and reinforcers or between actions and reinforcers when the value of the reinforcement shifts. In contrast, the roles of the right CMA and the left BLA are more specific, with the right CMA primarily involved in processing negative prediction errors during aversive classical learning, and the left BLA being responsible for encoding positive prediction errors during appetitive instrumental learning. These findings not only shed light on the functional organization of the amygdala in processing prediction errors in healthy individuals but also provide valuable insights into the neural mechanisms underlying conditions like obesity and eating disorders, which are characterized by impaired learning from prediction errors in contexts involving food and social cues.
Copyright-protected material. May be used within the limits of statutory user freedoms
Copyright holder:Publication made available with the written permission of the author
Digitizing institution:Nencki Institute of Experimental Biology of the Polish Academy of Sciences
Original in:Library of the Nencki Institute of Experimental Biology PAS
Access: