She started with a detailed overview of LTP formation, she narrowed down to LTP in NMDA receptors in CA1 region of the hippocampus. She discussed several early papers starting with the Morris water maze test in 1984 that supported the hypothesis that learning and memory could indeed be directly correlated to LTP. These papers demonstrated that blocking of NMDAR or some of its subunits (NR2B) lead to learning impairment.
She then discussed several later papers that show that there is no direct correlation. For example, severe early life stress (24 hour maternal separation in neonatal rodents) hampers learning and neurogenesis but improves hippocampal plasticity under high stress for adults. LTP is also seen in other areas (eg: addiction to cocaine, nicotine causes LTP in the VTA).
The current status of research seems to indicate that LTP is perhaps necessary but not sufficient in the memory formation and retrieval processes.
The seminar provoked plenty of lively interactions - some hypothetical questions, some comments targeted to the topic and on the various experimental approaches taken up in the papers.
"Does the brain prefer oscillatory stimuli or LTP type of stereotyped inputs?",
"Interpretation of data based on Gene knockouts needs to be done with care as there is significant impact on aspects of functioning other than the one under study."
"Isn't focus on NMDA too simplistic as it is involved in many other activities for normal functioning?"
"What about learning aspects outside the hippocampus - involving the cortices?"
Overall, great coverage of the topic and a stimulating debate on its different aspects.