Findings from lesion studies suggest the HPC is critical for memory of only certain types of information, such as places (specific locations in allocentric space) or contexts, and less important for remembering objects. But many tasks used to demonstrate content-specific memory deficits differ in ways other than just the type of information that must be remembered. Other differences include the nature of the motivation, the reinforcement contingencies, amount of training required, the behaviours that are measured, and the degree of arousal induced by the procedures. Any of these variables could potentially interact with the functional consequences of HPC damage in ways that produce a misleading picture of how well various types of information are remembered. By using variants of the novelty-preference test, we believe we can circumvent many of these confounds while comparing the effects of brain lesions on memory for contextual, allocentric-spatial, and object information.

We have tested rats with HPC lesions on these three different novelty-preference trials (context, place, and object-recognition). They displayed a normal preference on object trials, but not on place trials or context trials (Mumby et al., 2002). These results suggest that HPC damage has a disproportionate effect on memory for the environmental context in which particular objects are encountered, and the locations within the environment where they are encountered, with less effect on memory for the objects themselves. Also, an important feature of these three trial types is that the sample phase is identical for all three types -5 minutes in the arena with two identical objects. The subtle manipulation that sets them apart occurs on the retention test, so there is no confounding of memory for different types of information with the nature of the learning events. Moreover, the behaviour used to index memory is the same on all three versions. 

Testing Context Memory using a Variant of the Novelty Preference Test

This version assesses whether a rat remembers that a particular object was encountered in a particular context. Rats are familiarized with two different objects, each for 5 minutes, and each in a unique context (different open fields); exploratory preference is later measured on a test where both objects are presented in one of the two contexts. Normal rats spend more time exploring the object that is now in a context different from that during familiarization. In the Mumby et al., 2002 experiment, the rats with HPC lesions were impaired on the context trials; however, we did not know whether this impairment was due to an inability to remember the contexts, or was it because they did not remember which object appeared in which context. In order to address this question, we performed experiments using a context-shift type of novelty-preference test.

Typically, normal rats show a weakened preference for the novel object when the sample and test phases are conducted in different contexts, indicating that retrieval of contextual cues can facilitate recognition. We hypothesized that if rats with pretraining HPC lesions are unable to remember the context in which they’re exposed to during the learning phase, their performance should not be impaired when the context is different for the test phase. Our results showed that the control rats and HPC rats performed normally when the context remained the same for both the learning and test phase. However, when learning and testing occurred in two different contexts, rats with HPC lesions were impaired while control rats performed normally (O'Brien et al., 2006). The results suggested that rats with HPC lesions must be incorporating the context during learning considering they were impaired when the context was changed for the test phase. We hypothesized that the hippocampus must play an important role in recognizing previously learned objects appearing in subsequent contexts.

We recently conducted experiments similar to this one but using a variant of the open field apparatus, the circular track apparatus. In one experiment, rats were familiarized to objects in one circular track and the test phase was either conducted in the same circular track or a second circular track located in a different room. In this experiment, we found similar results to the O’Brien et al. experiment. In a second experiment, we wanted to determine how novel-object preference is affected by changing the local cues that surround the objects for the test phase. In order to test this, we used one circular track in one room. The compartments of the circular track were contextually distinct (local cues) while the context of the room (global cues) remained the same. On the test phase, half of the objects were moved to a different compartment and the other half remained in their respective compartment. The results showed that when the objects remained in the same compartment for the test phase, rats with HPC lesions and control rats were not impaired on the NOP test. However, when objects were moved to a different compartment for the test phase, rats with HPC lesions were not impaired while control rats were (Piterkin et al., 2008). The phenomena of response habituation and dishabituation may explain why the control rats failed to display a novel-object preference when the sample object was moved to a new place and paired with a novel object. When the sample object was encountered in a different place, the exploratory response that it evoked was dishabituated and the rat therefore explored the sample more than it would have if it hadn’t moved. Thus, the lack of novel-object preference when the object was moved to a different compartment shows that control rats were able to detect that the sample object changed location. According to this interpretation, the rats with HPC lesions showed a novel object preference in the different compartments because they could not remember where the sample object had been previously.

The findings from these experiments are consistent with the view that HPC damage does not cause a general inability to recognize objects, or an inability to encode or store representations of the environment in which these objects are encountered. The results suggest that HPC damage impairs the ability to remember the specific locations of familiar objects within a particular context.

Testing Place Memory using a Variant of the Novelty Preference Test

This version of the novelty-preference test assesses whether a rat recognizes that an object is in a place where there had not previously been an object. In the typical place memory task, a rat is familiarized with two identical objects. Following a delay, the rat is placed back into the apparatus, this time one of the objects is moved to a different location within the open field arena. Normal rats spend more time exploring the object that has moved.

Although this type of place-memory test has provided us with information on the effects of place memory, we have found inconsistent results. One reason for this inconsistency is possibly due to the small size of the open field arena. Recently, Dr. Gaskin designed a larger open field apparatus in order to test rats on their place memory. Now, the displacement of an object is more noticeable while at the same time, providing the rat with the ability to see extra-maze cues. These adjustments have allowed us to measure types of behaviours that are expressed that we couldn’t before (e.g., the amount of time a rat spends in the area where the object used to be). The procedure is similar to the typical place memory test: Rats are familiarized with two identical objects for a designated amount of time over several consecutive days. Following a delay, rats are placed back into the apparatus for the test phase. On the test phase, one object remains in its same location while the other object is moved to a new location within the circular open field.

In one experiment, rats were familiarized to two identical objects for 7 minutes a day for 7 consecutive days and then received either sham surgery or HPC lesions 1-3 days after the last familiarization session or 3 weeks later. Following recovery, rats were placed back into the apparatus, this time one of the objects was displaced to a new location while the other remained in its respective location. Rats that received HPC lesions 3 weeks after the last familiarization session showed evidence of preserved remote spatial memory, whereas the rats that received HPC lesions 1-3 days after familiarization did not (Gaskin, Tardif & Mumby, 2009). This suggests that the rats that received their HPC lesions 3 weeks after the last familiarization session, displayed evidence of temporally graded retrograde amnesia. Another interesting finding was the difference in the way in which the remote memory was expressed behaviourally between the HPC and control rats. Rats with HPC lesions spent more time investigating the displaced object versus the non-displaced one, whereas control rats spent more time in the quadrant where the displaced object used to be rather than the quadrant it was relocated to.

To learn more about Dr. Gaskin’s research please click HERE.

The Delayed Matching-to-Place (DMTP) task in the water maze assesses a rat's ability to retain, over various delays, information about where it recently found the hidden platform, on a single occasion. During the sample phase of each trial, the rat searches for and finds the hidden platform. After a delay, the rat is returned to the pool and allowed to find the platform again in the same location. Decreased escape latency from the first swim to the second swim reflects the extent to which the rat remembers the platform location. This task places a heavy demand on working-memory. We recently conducted an experiment using the water-maze to assess ischemic rats’ spatial memory. The results suggested that ischemic rats were impaired relative to control rats on the DMTP task when a 15 and 60 second delay was used.