File 'Reyetal_NatComm_dataset.mat' has the response dataset associated to the paper: 

Rey et al, "Encoding of long-term associations through neural unitization in the human medial temporal lobe", Nat Commun. 2018 Oct 22;9(1):4372. doi: 10.1038/s41467-018-06870-2.

There are 3 variables common to all the responses:

time_pre_stim = 1; number of seconds from stimulus onset that were epoched on each trial. Same value for LFP and spikes
time_pos_stim = 2; number of seconds after stimulus onset that were epoched on each trial. Same value for LFP and spikes
time_LFP_extra = 0.5; number of additional seconds on each LFP trial at the beginning and end of the epoched LFP. This is to avoid border effects when the LFP is further filtered

The structure 'resp' has all the responses, with the following fields:

1) sessionID. The session number in which the response was found. From 1 to 21.
2) stimulus_name. The name of the jpg file that was presented as a stimulus
3) recording_site. Area where the electrode was implanted. RAH = right anterior hippocampus, LAH = left anterior hippocampus, RAM = right amygdala, LAM = left amygdala, RPH = right posterior hippocampus, LPH = left posterior hippocampus.  
4) channel_number. Channel number in the recording setup.
5) cluster_number: Cluster number following spike sorting. Each cluster is associated to a different putative single neuron.
6) LFP_sr: Sampling rate for the LFP signal
7) original_sr: Original sampling rate used in the acquisition system
8) LFP: matrix where each row is associated to a trial and columns are samples. The samples on each trial are structured as: [ceil((time_LFP_extra+time_pre_stim)*LFP_sr),ceil((time_pos_stim+time_LFP_extra)*LFP_sr)].     
9) spike_stimes: cell array where each row contains a vector with the spike times associated to the occurrence of each spike, measured in msec for stimulus onset.