Examining Reject Relations in Stimulus Equivalence Simulations
This work addresses a methodological problem for researchers using computational models to study stimulus equivalence in psychology, highlighting that apparent equivalence in simulations might be artifactual.
This study investigated whether artificial neural networks (FFNs, BERT, GPT) demonstrate stimulus equivalence class formation or rely on associative learning in matching-to-sample simulations with reject relations. Results showed that while some agents achieved high accuracy on equivalence tests, their performance was comparable to a probabilistic associative benchmark, suggesting these models may use associative strategies rather than true equivalence.
Simulations offer a valuable tool for exploring stimulus equivalence (SE), yet the potential of reject relations to disrupt the assessment of equivalence class formation is contentious. This study investigates the role of reject relations in the acquisition of stimulus equivalence using computational models. We examined feedforward neural networks (FFNs), bidirectional encoder representations from transformers (BERT), and generative pre-trained transformers (GPT) across 18 conditions in matching-to-sample (MTS) simulations. Conditions varied in training structure (linear series, one-to-many, and many-to-one), relation type (select-only, reject-only, and select-reject), and negative comparison selection (standard and biased). A probabilistic agent served as a benchmark, embodying purely associative learning. The primary goal was to determine whether artificial neural networks could demonstrate equivalence class formation or whether their performance reflected associative learning. Results showed that reject relations influenced agent performance. While some agents achieved high accuracy on equivalence tests, particularly with reject relations and biased negative comparisons, this performance was comparable to the probabilistic agent. These findings suggest that artificial neural networks, including transformer models, may rely on associative strategies rather than SE. This underscores the need for careful consideration of reject relations and more stringent criteria in computational models of equivalence.