Episodic memories, or memories for where/what/when information, is a dynamic process. Whenever we learn or encode information, it takes time and overnight sleep to stabilize and consolidate such information. When such consolidated memories are remembered, they become active in the brain and can be modified by new learning. Such updating of memories is an ongoing process and helps link new learning to prior knowledge.
The brain region primarily important for forming and stabilizing memories is the hippocampus, seated deep inside the brain behind our ears. This region helps to strengthen the connections between memory traces laid out across the brain. The hippocampus coordinates with the prefrontal cortex to organize and integrate new information with the memory networks. This process can days to weeks to months.
The structure and function of the hippocampus and prefrontal cortex are known to be affected in depression, which results in memory impairment. However, studies on memory in depression have mostly tested recall performance in the same session, a few minutes to up to an hour after learning. We lack studies that test memory over multiple days in depressed individuals. We, therefore, set out to test this.
Unipolar and bipolar depression in-patients admitted to the National Institute of Mental Health and Neurosciences (NIMHANS) were recruited. Age- and educationally-matched, healthy individuals from the community formed the control group. All participants learned a set of objects on Day 1. Forty-eight hours later, on Day 3, half the participants were reminded of what they did on Day 1 while the other half were not. Both reminded and non-reminded participants then learned a different set of objects. After another two days, on Day 5, all participants had to recall objects from Day 1.
During recall, four days later, on Day 5, both unipolar and bipolar depression patients remembered fewer objects than control participants. This impairment was expected and in line with previous studies of memory in depression.
Control participants who were reminded of their learning experience from Day 1 also recalled objects from Day 3. Control participants who were not reminded did not recall objects from Day 3. This is typically known as reactivation-dependent memory updating. Only when Day 1 memory was reactivated, before learning a different set objects on Day 3, did the Day 1 memory get updated with objects from Day 3. Such updating does not occur in the non-reminded control participants.
This reactivation-dependent memory updating was not affected in unipolar depression patients. Their memory updating performance was similar to the control participants. However, bipolar depression patients showed an unexpected pattern of recall. Irrespective of the reminder, both reminded and non-reminded bipolar depression patients recalled objects from both Day 1 and Day 3.
We confirmed this in a second experiment where all participants had to recall objects from Day 3 on Day 5. Only bipolar depression patients recalled objects from Day 1 along with objects from Day 3. Control participants and unipolar patients recalled objects only from Day 3. In other words, Day 3 memory is not reactivated once it is learned, therefore it does not get updated with Day 1 objects, as seen in control participants and unipolar depression patients. Yet, bipolar depression patients recalled objects from both Day 1 and Day 3, suggesting indiscriminate recall.
Thus, unipolar depression patients were able to selectively update only the reactivated memory, whereas patients with bipolar depression were recalling indiscriminately, irrespective of memory reactivation. This inability to remember which object belongs to which list is known as source confusion. This phenomenon points to what we see in bipolar depression.
These results are relevant in the light of therapies that aim to overwrite traumatic memories in psychiatric conditions.
These findings are described in the article entitled Differential effects of unipolar versus bipolar depression on episodic memory updating, recently published in the journal Neurobiology of Learning and Memory.