Carhart-Harris RL, Feilding A, Wise RG, Nutt DJ et al. British Journal of Psychiatry, 2012, 200(3):238–44.
This study, carried out as part of the Beckley-Imperial Psychedelic Research Programme, aimed to investigate the hypothesis that psilocybin facilitates access to personal memories and emotions by comparing the subjective and neural response to positive autobiographical memory recollection under psilocybin versus placebo. 
(See below for METHOD)
FINDINGS and IMPLICATIONS
- Under psilocybin greater activations were seen in sensory cortical regions in the late phase of memory recollection. Under placebo, these regions showed deactivations.
- Participants rated the memories as more vivid and visual under psilocybin.
- Late phase right parahippocampal activations correlated positively with how emotional the memories were.
- In a follow-up study the vividness of the memories correlated positively with their later subjective wellbeing.
- 15 healthy participants recollected positive autobiographical memories under the influence of psilocybin or placebo during fMRI scanning.
- The two scanning sessions were carried out in balanced order (some participants received placebo first, others psilocybin) and separated by approximately seven days.
- The participants were asked to generate at least 30 personal memories of specific positive autobiographical events. They were also asked to generate a cue sentence to prompt recall of each memory (which they were free to encrypt or code to keep confidential). Lastly, participants estimated their age when each recollected event took place, and rated the emotional potency of the memory.
- Using this information, the memories were split into two sets of 15, balanced for recency and potency, and then randomly assigned to either the psilocybin or placebo conditions for each participant.
- Psilocybin or placebo was administered intravenously. In the case of psilocybin, this ensures that a small dose rapidly induces its subjective and neural effects.
- 3 different conditions were repeated 15 times in the same order during both scanning sessions:
- Rest: Auditory tone followed by visual instruction ‘relax’ (presented for 6 seconds) then ‘close your eyes’. Another auditory tone 16s later signalled for participants to open their eyes. This gave an unbiased baseline, matched for sensory input (i.e. eyes closed) to compare recollection to.
- Memory task: A memory cue sentence was presented (6s) before the instruction ‘close your eyes’. 16s was given for participants to recollect before another auditory tone, signalling for them to open their eyes.
- Attention task: Visual instruction to ‘count the beeps’ (6s) followed by ‘close your eyes’ and onset of attention task (16s). The chosen task was a sequence of auditory tones of different pitch playing in left or right ear, with one extra tone in either left or right – participants had to identify which ear heard more tones. This task prevents carry over effects from memory trials into rest trials, by distracting participants from continued recollection.
- Memory ratings (0-10 scale) of vividness, emotional intensity, valence and the extent to which the memory elicited visual imagery were given by participants (e.g. ‘How strong was the emotion you felt, with 10 being extremely intense?’) when they exited the scanner.
- Follow-up ratings in subjective well-being and life satisfaction were taken two weeks later on scale of -3 (decreased very much) to +3 (increased very much) with 0 as ‘no change’.