Dendritic spines and memory formation in the chick

Patel, Sanjay N (1989). Dendritic spines and memory formation in the chick. PhD thesis The Open University.



Passive avoidance training results in a number of biochemical, morphological, and electrophysiological changes in the forebrain of the one day-old chick. One particular region in which these alterations occur is the intermediate part of the medial hyperstriatum ventrale (IMHV). This thesis reports several morphological experiments on the effects of passive avoidance training on dendritic spines and dendritic branching patterns of large, multipolar, projection neurons (abbreviated LMPNs) in the IMHV, in order to determine structural correlates of memory formation for this task, at the light microscope level.

The chicks were trained on a passive avoidance training task by presentation of a shiny chrome bead coated with a bitter tasting substance (methylanthranilate). The chicks will spontaneously peck at the bead, but show a characteristic disgust response on the first peck and avoid a similar (dry) bead subsequently (trained or M-chicks). Control chicks were presented with a water coated bead, which they do not find aversive (W-chicks). 24-26h later, the chicks were perfused with an aldehyde fixative, and the left and right IMHV regions were dissected out and Golgi-impregnated by the rapid Golgi method. The blocks were sectioned with a tissue chopper at 90-120p.m, and permanently mounted onto slides, in DPX, which were coded so that subsequent procedures were performed "blind".

LMPNs in the left and right IMHV regions from trained and control chicks were identified and examined for changes in spine density, spine shape and in dendritic branching patterns (which were analysed by the Sholl concentric ring method, by the number of dendrites at each br^ch order and by vertex analysis).

Training resulted in (1) a significant increase in spine density in the left and right hemispheres and (2) an increase in the mean diameter of the spine heads with concomitant shortening of the spine stems, but only significantly in. the left hemisphere. These changes occured without a significant change in the mean overall spine length and also without significant alterations in the lengths or the diameters of the dendrite branches. A significant hemispheric asymmetry was also observed between the left (L) and right (R) hemispheres of control chicks: R>L, but no asymmetry was found in trained chicks.

No significant differences in branching patterns were found after passive avoidance training.

Because the changes in spine density and shape may have been caused by nonspecific factors associated with the training experience, such as stress, arousal, or the taste of the methyl anthranilate, per se, a further experiment was conducted in which trained chicks were given a brief, subconvulsive, trahscranial electroshock, 5 min after showing the disgust response. This rendered approximately half of the chicks amnesic, the rest showed recall when tested 24h after training. Control chicks receiving the same treatment showed no change in pecking behaviour. A significant increase in spine density was found in the recall group compared with either the amnesic or the shocked water control groups, but only in the left hemisphere. However, no alterations in spine shape and no significant differences in dendrite lengths or dendrite diameters were noted.

In conclusion: (1) passive avoidance training is associated with an increase in spine density that is specifically related to long-term memory formation for the task; (2) spine density and shape changes can occur within 24h of a single-trial learning experience; (3) these alterations occur without significant differences in branching patterns of the LMPNs, suggesting either that it may take longer than 24h for observable changes in branching patterns to occur, or that they may not be involved in memory storage for the training task, and (4) because the shape changes were only found on LMPNs in the left hemisphere of trained chicks, this hemisphere may be predominantly involved in the memory storage processes for this task. This is also supported by qualitative results from the dendritic branching investigation.

Viewing alternatives

Download history


Public Attention

Altmetrics from Altmetric

Number of Citations

Citations from Dimensions

Item Actions