This project was funded in part from a grant from the National Science Foundation (D.b.-A.) (PHY0555312).
An artificial neural network model of orienting attention toward threatening somatosensory stimuli
Version of Record online: 26 OCT 2007
Copyright © 2007 Society for Psychophysiological Research
Volume 45, Issue 2, pages 229–239, March 2008
How to Cite
Dowman, R. and Ben-Avraham, D. (2008), An artificial neural network model of orienting attention toward threatening somatosensory stimuli. Psychophysiology, 45: 229–239. doi: 10.1111/j.1469-8986.2007.00614.x
- Issue online: 26 OCT 2007
- Version of Record online: 26 OCT 2007
- (Received May 2, 2007; Accepted August 17, 2007)
- Artificial neural network;
- Medial prefrontal cortex
An artificial neural network model was designed to test the threat detection hypothesis developed in our experimental studies, where threat detector activity in the somatosensory association areas is monitored by the medial prefrontal cortex, which signals the lateral prefrontal cortex to redirect attention to the threat. As in our experimental studies, simulated threat-evoked activations of all three brain areas were larger when the somatosensory target stimulus was unattended than attended, and the increase in behavioral reaction times when the target stimulus was unattended was smaller for threatening than nonthreatening stimuli. The model also generated a number of novel predictions, for example, the effect of threat on reaction time only occurs when the target stimulus is unattended, and the P3a indexes prefrontal cortex activity involved in redirecting attention toward response processes on that trial and sensory processes on subsequent trials.