The Hunt for the Memory Engram

The Hunt for the Memory Engram

1. History: What parts of the brain are involved in memory?

   • Karl Lashley (1915-50)
     • Mass Action: all parts of the cortex contribute to support cognition in memory function.
     • Equipotentiality: all parts of the cortex contribute equally to memory function.
   • Von Bechtrew
     • In patients who have memory problems, there was a very distinct area of the brain showing abnormalities - the temporal lobe.
     • People who have memory difficulties have softening (neural damage) of the temporal lobe
     • The first person to identify that the temporal lobe may have something to do with memory.
   • Patient H.M.
     • By age of 20, he was suffering from Mal seizures, making him unable to work.
     • In 1953 (age 29) he underwent bilateral temporal lobe surgery - removed brain regions causing the seizures.
     • Consequences of the surgery for memory had not been considered
     • Since lesion, H.M. suffered from severe anterograde amnesia - marked deficit for every day events. He was unable to encode new info about events that occurred in his life.
     • Also showed a temporally gradient retrograde amnesia - a loss of memory for events that occurred before the surgery (memory for events occurring 5 or 6 years prior to the surgery were affected). This was supported by his family members - He could remember events in his childhood and adolescents.
     • Association between acquisition of new memories and the relative sparing of old memories - implies that the damage caused appears to be responsible for encoding, and storing (At least for a time) memory for events that happened to him. But memories eventually became independent of the damage inflicted on the temporal lobe.
     • He kept an intact working (S-T) memory: normal digit span and showed rate of forgetting within a normal range providing he was able to constantly rehearse the info; If active rehearsal was interrupted, he would forget; Transfer from STM to LTM was clearly impaired.
     • However, his non-verbal STM was impaired
     • He had intact procedural memory - could learn new motor tasks (Mirror tracing task; Pursuit rotor tracking; Implicitly familiar with testing equipment).

2. Declarative versus Procedural Memory Systems

   • Squire & Tulving: there may be different memory systems supported by different brain regions - but when you lesion the temporal lobe, you lose the ability to form declarative memory (memories which you can verbally declare).
     • An important aspect of the declarative memory is episodic memory - for H.M., this declarative memory was lost; what was spared was procedural memory.

3. Animal model of human amnesia: non-matching-to-sample task in monkeys

   • The temporal lobe is made of a number of structures (e.g. hippocampus)
     • which of these structures are critical for declarative memory? - can only study these using animal models, because humans do not have selective brain damage.
   • Mortimer Mishkin (1978): Used a task that effectively was supposed to mirror the tasks used in amnesic patients. Recognition memory task/ non-matchhiing-to-sample task where monkeys were shown an object. It displaced this object and received a reward. After delay, the animal was shown a copy of the object and a novel objet - to get another reward animal was displace the novel object. Used normal monkeys compared to monkeys who underwent different lesions. In monkeys, when you only lesion the hippocampus, there is almost no deficit in memory (contrasting to previous beliefs).
   • Nemanic et al 2004: non-matching-to-sample task on rats. Damage to hippocampus had no effect on recognition memory. If you damage the cortical tissue, in particular the perirhinal cortex, then you get a clear deficit in these tests (in H.M. the cortical tissue was also damaged).
   • Work on animals has then shown that damage to the hippocampus alone does not disrupt a core feature of the amnesic syndrome in humans, that is recognition memory. But you can get profound recognition deficits by lesioning the cortical tissue.

4. Dual process model of recognition memory. John Aggelton: Familiarity vs Recall

   • Disruption to the hippocampus does impact one core feature of memory: spatial memory - the ability to remember the location that you were in.
   • Idea that there are different types of recognition memory: relies on a sense of familiarity; involves the recall of the spatial and temporal context in which that item was presented - episodic version of recognition.
   • Proposed that the hippocampus is important for recognition memory that involved the binding of space & time. He further suggested that the perirhinal cortex is only really invoice in recognition relying on familiarity. Tests in animals only test the familiarity version of recognition memory - the hippocampus is not necessary.