Local Synaptic Modification Can Lead to Organized Connectivity Patterns in Associative Memory

Author(s):  
Günther Palm
1993 ◽  
Vol 5 (1) ◽  
pp. 32-44 ◽  
Author(s):  
Michael E. Hasselmo

Implementing associative memory function in biologically realistic networks raises difficulties not dealt with in previous associative memory models. In particular, during learning of overlapping input patterns, recall of previously stored patterns can interfere with the learning of new patterns. Most associative memory models avoid this difficulty by ignoring the effect of previously modified connections during learning, thereby clamping activity to the patterns to be learned. Here I propose that the effects of acetylcholine in cortical structures may provide a neuropsychological mechanism for this clamping. Recent brain slice experiments have shown that acetylcholine selectively suppresses excitatory intrinsic fiber synaptic transmission within the olfactory cortex, while leaving excitatory afferent input unaffected. In a computational model of olfactory cortex, this selective suppression, applied during learning, prevents interference from previously stored patterns during the learning of new patterns. Analysis of the model shows that the amount of suppression necessary to prevent interference depends on cortical parameters such as inhibition and the threshold of synaptic modification, as well as input parameters such as the amount of overlap between the patterns being stored.


2006 ◽  
Vol 69 (10-12) ◽  
pp. 1305-1308 ◽  
Author(s):  
Junmei Zhu ◽  
Christoph von der Malsburg

1994 ◽  
Vol 72 (2) ◽  
pp. 659-677 ◽  
Author(s):  
E. Barkai ◽  
R. E. Bergman ◽  
G. Horwitz ◽  
M. E. Hasselmo

1. Associative memory function was analyzed in a realistic biophysical simulation of rat piriform (olfactory) cortex containing 240 pyramidal cells and 58 each of two types of inhibitory interneurons. Pyramidal cell simulations incorporated six different intrinsic currents and three different synaptic currents. We investigated the hypothesis that acetylcholine sets the appropriate dynamics for learning within the network, whereas removal of cholinergic modulation sets the appropriate dynamics for recall. The associative memory function of the network was tested during recall after simulation of the cholinergic suppression of intrinsic fiber synaptic transmission and the cholinergic suppression of neuronal adaptation during learning. 2. Hebbian modification of excitatory synaptic connections between pyramidal cells during learning of patterns of afferent activity allowed the model to show the basic associative memory property of completion during recall in response to degraded versions of those patterns, as evaluated by a performance measure based on normalized dot products. 3. During learning of multiple overlapping patterns of afferent activity, recall of previously learned patterns interfered with the learning of new patterns. As more patterns were stored this interference could lead to the exponential growth of a large number of excitatory synaptic connections within the network. This runaway synaptic modification during learning led to excessive excitatory activity during recall, preventing the accurate recall of individual patterns. 4. Runaway synaptic modification of excitatory intrinsic connections could be prevented by selective suppression of synaptic transmission at these synapses during learning. This allowed effective recall of single learned afferent patterns in response to degraded versions of those patterns, without interference from other learned patterns. 5. During learning, cholinergic suppression of neuronal adaptation enhanced the activity of cortical pyramidal cells in response to afferent input, compensating for decreased activity due to suppression of intrinsic fiber synaptic transmission. This modulation of adaptation led to more rapid learning of afferent input patterns, as demonstrated by higher values of the performance measure. 6. During recall, when suppression of excitatory intrinsic synaptic transmission was removed, continued cholinergic suppression of neuronal adaptation led to the spread of excessive activity. More stable activity patterns during recall could be obtained when the cholinergic suppression of neuronal adaptation was removed at the same time as the cholinergic suppression of synaptic transmission. 7. A realistic biophysical simulation of the effects of acetylcholine on synaptic transmission and neuronal adaptation in the piriform cortex shows that these effects act together to set the appropriate dynamics for learning, whereas removal of both effects sets the appropriate dynamics for recall.


2006 ◽  
Author(s):  
Yvonne Wakeford ◽  
Michael T. Carlin ◽  
Michael P. Toglia

Sign in / Sign up

Export Citation Format

Share Document