Semantic integration by pattern priming: experiment and cortical network model

Abstract

Neural network models describe semantic priming effects by way of mechanisms of activation of neurons coding for words that rely strongly on synaptic efficacies between pairs of neurons. Biologically inspired Hebbian learning defines efficacy values as a function of the activity of pre-and post-synaptic neurons only. It generates only pair associations between words in the semantic network. However, the statistical analysis of large text databases points to the frequent occurrence not only of pairs of words (e.g., ''the way'') but also of patterns of more than two words (e.g., ''by the way''). The learning of these frequent patterns of words is not reducible to associations between pairs of words but must take into account the higher level of coding of three-word patterns. The processing and learning of pattern of words challenges classical Hebbian learning algorithms used in biologically inspired models of priming. The aim of the present study was to test the effects of patterns on the semantic processing of words and to investigate how an inter-synaptic learning algorithm succeeds at reproducing the experimental data. The experiment manipulates the frequency of occurrence of patterns of three words in a multiple-paradigm protocol. Results show for the first time that target words benefit more priming when embedded in a pattern with the two primes than when only associated with each prime in pairs. A biologically inspired inter-synaptic learning algorithm is tested that potentiates synapses as a function of the activation of more than two pre-and postsynaptic neurons. Simulations show that the network can learn patterns of three words to reproduce the experimental results.

Statistical learning of unbalanced exclusive-or temporal sequences in humans

“…This ability to extract the combination of two preceding items to better predict the following one takes the form of a quite complex computational representation of probability values. Such learning of second-order TP challenges our understanding of combinatorial learning [ 32 , 33 ]. Biologically inspired models of learning encode transitional probability through the efficacy of synapses [ 55 , 56 ].…”

“…The experimental design consisted of four triplets of the form X − Y − Z that combined items according to an XOR logic [ 33 ]. The XOR allows us to dissociate between first-order TPs ( p ( Y | X ), p ( Z | Y ), and p ( Z | X )) and second-order TPs p ( Z | XY ).…”

“…The learning of TPs involves patterns that can occur in different sequences, with the consequence that the TP of an item given a preceding one ( p ( Y | X )) can be different from the frequency of the XY sequence. The reason is that the frequency of all sequences embedding X and Y need to be taken into account globally [ 32 , 33 ]. Indeed, if an item X occurs in several sequences (e.g., XY and XT ), the frequency of the sequence XY does not correspond to the TP p ( Y | X ).…”

“…The aim of the present study was therefore to investigate to what extent prediction in humans predominantly involves frequency, first-order TP or second-order TP during learning. To this end, we used repeated presentation of four sequences of three items combined according to the exclusive-or (XOR) logic [ 32 , 33 , 44 ].…”

Spontaneous analogising caused by text stimuli in design thinking: differences between higher- and lower-creativity groups

“…In particular, Kambe et al (2015) suggested that beta phase resetting occurred in areas related to the subsequent stimulus, supporting the idea that the perception of multisensory stimuli is simultaneous. Neural network models have been argued to describe semantic priming effects by activating neurons that code for words that rely strongly on synaptic efficacies between pairs of neurons (Lavigne et al 2016).…”