Context effects in auditory implicit memory

CINET: A Brain-Inspired Deep Learning Context-Integrating Neural Network Model for Resolving Ambiguous Stimuli

The brain uses contextual information to uniquely resolve the interpretation of ambiguous stimuli. This paper introduces a deep learning neural network classification model that emulates this ability by integrating weighted bidirectional context into the classification process. The model, referred to as the CINET, is implemented using a convolution neural network (CNN), which is shown to be ideal for combining target and context stimuli and for extracting coupled target-context features. The CINET parameters can be manipulated to simulate congruent and incongruent context environments and to manipulate target-context stimuli relationships. The formulation of the CINET is quite general; consequently, it is not restricted to stimuli in any particular sensory modality nor to the dimensionality of the stimuli. A broad range of experiments is designed to demonstrate the effectiveness of the CINET in resolving ambiguous visual stimuli and in improving the classification of non-ambiguous visual stimuli in various contextual environments. The fact that the performance improves through the inclusion of context can be exploited to design robust brain-inspired machine learning algorithms. It is interesting to note that the CINET is a classification model that is inspired by a combination of brain's ability to integrate contextual information and the CNN, which is inspired by the hierarchical processing of information in the visual cortex.

The effect of generation on long-term repetition priming in auditory and visual perceptual identification

Perceptual implicit memory is typically most robust when the perceptual processing at encoding matches the perceptual processing required during retrieval. A consistent exception is the robust priming that semantic generation produces on the perceptual identification test (Masson & MacLeod, 2002), a finding which has been attributed to either (1) conceptual influences in this nominally perceptual task, or (2) covert orthographic processing during generative encoding. The present experiments assess these possibilities using both auditory and visual perceptual identification, tests in which participants identify auditory words in noise or rapidly-presented visual words. During the encoding phase of the experiments, participants generated some words and perceived others in an intermixed study list. The perceptual control condition was visual (reading) or auditory (hearing), and varied across participants. The reading and hearing conditions exhibited the expected modality-specificity, producing robust intra-modal priming and non-significant cross-modal priming. Priming in the generate condition depended on the perceptual control condition. With a read control condition, semantic generation produced robust visual priming but no auditory priming. With a hear control condition, the results were reversed: semantic generation produced robust auditory priming but not visual priming. This set of results is not consistent with a straightforward application of either the conceptual-influence or covert-orthography account, and implies that the nature of encoding in the generate condition is influenced by the broader list context.