Computational modeling of the negative priming effect based on inhibition patterns and working memory

Area of Science:

• Cognitive Psychology
• Computational Neuroscience
• Human Reaction Time Studies

Background:

• Negative priming (NP) describes response slowing to previously ignored stimuli, observed in paradigms like the Stroop task.
• Existing theories suggest NP involves attention, memory, and inhibition, but a unified account is lacking due to stimulus complexity.
• Understanding NP subtypes is crucial for a comprehensive theoretical framework.

Purpose of the Study:

• To investigate varying degrees of the negative priming effect based on prime-probe combinations within the Stroop task.
• To explore the roles of selective attention and working memory in generating differential NP effects.
• To computationally model NP effects to elucidate underlying cognitive mechanisms.

Main Methods:

• Recorded reaction times (RTs) from 66 healthy participants performing a classic Stroop task.
• Analyzed three NP subtypes defined by prime-probe pair configurations.
• Employed a parallel distributed processing (PDP) model and a modified PDP model with working memory (PDP-WM) for computational analysis.

Main Results:

• Significant differences in RTs were observed among NP subtypes, attributed to differential disinhibition (release from inhibition).
• The PDP-WM model, incorporating working memory, successfully simulated the observed differential NP effects, unlike the conventional PDP model.
• Behavioral data closely matched the simulation outcomes from the PDP-WM model.

Conclusions:

• Working memory plays a critical role in the re-accumulation of evidence for target responses, leading to differential NP effects.
• The PDP-WM model provides a better fit for behavioral data, suggesting its utility in understanding NP mechanisms.
• These findings contribute to an integrated theoretical account of complex negative priming effects.