Math anxiety and associative knowledge structure are entwined in psychology students but not in Large Language Models like GPT-3.5 and GPT-4o

TL;DR

This study uses behavioural forma mentis networks (BFMNs) to map how psychology students differently structure associative knowledge and affect regarding math and anxiety, and compares these human patterns with GPT-3.5 and GPT-4o simulations. In humans, the network prominence of the concept 'anxiety' (degree) and its negative valence reliably predict math anxiety across multiple facets, while 'math' itself shows weaker links to anxiety. GPT models, especially GPT-3.5, show weaker or qualitatively different links, with GPT-4o partially mirroring human patterns but still lacking the depth of affective grounding seen in people. The findings highlight a fundamental limit of current LLMs as affective cognitive simulators and point to BFNM-based metrics as informative tools for education and intervention, while emphasizing the enduring role of teachers in anxiety-sensitive learning support.

Abstract

Math anxiety poses significant challenges for university psychology students, affecting their career choices and overall well-being. This study employs a framework based on behavioural forma mentis networks (i.e. cognitive models that map how individuals structure their associative knowledge and emotional perceptions of concepts) to explore individual and group differences in the perception and association of concepts related to math and anxiety. We conducted 4 experiments involving psychology undergraduates from 2 samples (n1 = 70, n2 = 57) compared against GPT-simulated students (GPT-3.5: n2 = 300; GPT-4o: n4 = 300). Experiments 1, 2, and 3 employ individual-level network features to predict psychometric scores for math anxiety and its facets (observational, social and evaluational) from the Math Anxiety Scale. Experiment 4 focuses on group-level perceptions extracted from human students, GPT-3.5 and GPT-4o's networks. Results indicate that, in students, positive valence ratings and higher network degree for "anxiety", together with negative ratings for "math", can predict higher total and evaluative math anxiety. In contrast, these models do not work on GPT-based data because of differences in simulated networks and psychometric scores compared to humans. These results were also reconciled with differences found in the ways that high/low subgroups of simulated and real students framed semantically and emotionally STEM concepts. High math-anxiety students collectively framed "anxiety" in an emotionally polarising way, absent in the negative perception of low math-anxiety students. "Science" was rated positively, but contrasted against the negative perception of "math". These findings underscore the importance of understanding concept perception and associations in managing students' math anxiety.

Figures (9)

• Figure 1: Correlogram Experiment 1. The valence, degree and closeness centrality (CC) for the nodes "anxiety" and "math" are reported, along with the factors of MA: Evaluation, Everyday/Social (Social) and Passive Observation (Observation), together with the Total MA.
• Figure 2: Correlogram Experiment 2. The valence, degree and closeness centrality (CC) for the nodes "anxiety" and "math" are reported, along with the factors of MA: Evaluation, Everyday/Social (Social) and Passive Observation (Observation), together with the total MA.
• Figure 3: Correlogram Experiment 3 - GPT-3.5. The valence, degree and closeness centrality (CC) for the nodes "anxiety" and "math" are reported, along with the factors of MA: Evaluation, Everyday/Social (Social) and Passive Observation (Observation), together with the total MA.
• Figure 4: Correlogram Experiment 3 - GPT-4o. The valence, degree and closeness centrality (CC) for the nodes "anxiety" and "math" are reported, along with the factors of MA: Evaluation, Everyday/Social (Social) and Passive Observation (Observation), together with the total MA.
• Figure 5: Semantic frames for the node "mathematics".