Stereotypes are well-received social psychological phenomena, yet there is still a lack of in-depth understanding of the representation structure of different stereotypes. Within the connectionism framework, the present study investigated the directionality of activation with food gender stereotypes, via the new mouse tracking paradigm which can reflect participants' real-time cognitive processes during responses.
Two experiments were conducted. Experiment 1 examined the priming effect of different types of food on gender groups. The results of Experiment 1 showed that the food category had a significant priming effect on the perception of the gender group, reflecting an Attribute → Group activation directionality. Experiment 2 included three sub-experiments using the same experimental paradigm but different levels of gender priming materials to explore the priming effect of gender groups on food categories According to the results, the concept of the gender group could not trigger the association of the food category.
In conclusion, based on the results of two experiments, the food gender stereotype is likely to be a unidirectional (Attribute → Group) representation structure. We suggest this particular finding can be explained within the connectionist theoretical framework that the different levels of crowdedness (i.e. different associated concepts in numbers) and weights (i.e. different salience and familiarity) of gender group and food category are the reasons why the food gender stereotype can only be activated unidirectionally (Attribute → Group). 

 

Neural computational models are essential research methods for in-depth analysis of neural information encoding processes. In recent years, the normalization model has emerged as a novel neural computational approach and has found wide application in cognitive activities. The model posits that the response of neurons sensitive to a stimulus is suppressed by other neurons, with the computation involving a ratio between the response of an individual neuron and the summed activity of a neuron pool. While initially proposed to elucidate neuronal responses in the primary visual cortex, the normalization model has found extensive use in the realms of sensory perception and attention, yet its application to higher cognitive activities has been limited. Future studies exploring the potential of normalization models in more advanced cognitive domains would be a valuable avenue for research.