Abstract

Latent inhibition (LI) refers to the retardation of learning about a stimulus following repeated non-reinforced exposure, whereas perceptual learning (PL) reflects enhanced discriminability among preexposed stimuli. Although both phenomena are fundamental to associative learning theories, they are usually studied separately, limiting direct comparison and obscuring shared mechanisms. This thesis developed and validated a novel within-subject paradigm for examining LI and PL under common experimental conditions, with the goal of establishing a behavioral platform for future neuroscience and translational research. In Chapter 1, we manipulated preexposure duration and found that extensive preexposure (12 days) yielded robust LI, while limited preexposure (4 days) produced PL, showing that both effects can emerge within the same design. Chapter 2 applied this approach to a within-modality context. Here, LI was attenuated and variable across individuals, suggesting that perceptual similarity between preexposed and novel cues reduced discrimination. Chapter 3 tested whether LI arises from attentional decrement or associative interference by aligning preexposure and test contingencies. No net LI or facilitation was observed, indicating that attentional and associative processes may interact in complex, cancelling ways. Together, these findings highlight the utility of a within-subject paradigm for investigating LI and PL, while also revealing boundary conditions related to preexposure amount, cue modality, and theoretical mechanism. This work establishes a foundation for probing the neural circuit underlying LI, including the hippocampus, amygdala, striatum, thalamus, prefrontal cortex, and dopaminergic midbrain regions, and offers a translational tool for studying attentional filtering deficits in disorders such as schizophrenia.