We're hoping that you might be able to shed some light on a few issues trying to set up the non-linear model in the PsPM v4.0.1 toolbox.
In trying to problem solve, we have simulated SCR fear conditioning data for 10 subjects using actual event
consist of 80 trials per subject (30 CS+reinforced, 10 CS+unreinforced, 40 CS-). The data were convolved with the HRF based on condition (2x CS+r, 1.5x CS+u, 0.5x CS-). In the experimental paradigm, the CS+
is present for 2.5s and there is a 1s delay from the offset of the CS+ to the onset of the US, such that the shock arrives 3.5s after the onset of the CS+.
We expected to see a significant fixed response amplitude for shock vs non-shock trials, however it appears that the fixed response values are all negative and that the flexible amplutitude is positive. Additionally, when looking at the conditions individually (CS+r contrasts = 1, all other contrasts = 0), the means for the fixed response are around 0. I'm not sure how to interpret these results and would appreciate any insight. I've attached stats and figures as well as the data and events files for one subject and the group level result.
Thank you,
Daria
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
my suspicion is that the simulated data are too different from what SCR data would look like. You write you used an HRF for simulations - is that a hemodynamic response function? This would have a different delay and spread than SCR. You can easily see this by plotting your HRF together with an SCRF, e.g. the one implemented in PsPM (pspm_bf_scrf_f).
If you simulate data with this SCRF, then the model should fit (almost) perfectly. (Almost, because the non-linear model uses a slightly different RF). You could then check for robustness by perturbing the SCRF parameters.
Hope this helps
Dominik
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Thanks so much for your response. I've gone ahead and convolved the simulated data for a subset of subjects with the SCRF as you mentioned. The model does fit nearly perfectly. However, I'm still unsure about the interpretation of the results. The result of US+ minus US- still has a negative mean. As does US+ minus CS-. Further, the mean and sem of both are nearly identical. Lastly, when looking at individual conditions (CS+r contrasts = 1, all other contrasts = 0), the means for the fixed response are 0 for CS+r and CS+u.
I am seeing a similar pattern with my real data, and can't help but wonder if maybe the delay in shock onset after the offset of my CS+ could be responsible (the CS+ is present for 2.5s and there is a 1s delay from the offset of the CS+ to the onset of the US, such that the shock arrives 3.5s after the onset of the CS+). In both real and simulated results, the flexible response results seem to make more sense than the fixed.
Any insights on this would be great.
Thank you,
Daria
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I understand. This is odd indeed. I agree that the temporal gap between CS and US could have an impact - it is possible that there is a response to the CS offset which is not properly modelled. If you want, I'd be happy to look at your new simulations and results, and it could be useful to look at the mean time course for the different conditions in your experiment. If you don't want to post this publicly, send me an email
Best
Dominik
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
We're hoping that you might be able to shed some light on a few issues trying to set up the non-linear model in the PsPM v4.0.1 toolbox.
In trying to problem solve, we have simulated SCR fear conditioning data for 10 subjects using actual event
consist of 80 trials per subject (30 CS+reinforced, 10 CS+unreinforced, 40 CS-). The data were convolved with the HRF based on condition (2x CS+r, 1.5x CS+u, 0.5x CS-). In the experimental paradigm, the CS+
is present for 2.5s and there is a 1s delay from the offset of the CS+ to the onset of the US, such that the shock arrives 3.5s after the onset of the CS+.
We expected to see a significant fixed response amplitude for shock vs non-shock trials, however it appears that the fixed response values are all negative and that the flexible amplutitude is positive. Additionally, when looking at the conditions individually (CS+r contrasts = 1, all other contrasts = 0), the means for the fixed response are around 0. I'm not sure how to interpret these results and would appreciate any insight. I've attached stats and figures as well as the data and events files for one subject and the group level result.
Thank you,
Daria
Attachements
Hi Daria
my suspicion is that the simulated data are too different from what SCR data would look like. You write you used an HRF for simulations - is that a hemodynamic response function? This would have a different delay and spread than SCR. You can easily see this by plotting your HRF together with an SCRF, e.g. the one implemented in PsPM (pspm_bf_scrf_f).
If you simulate data with this SCRF, then the model should fit (almost) perfectly. (Almost, because the non-linear model uses a slightly different RF). You could then check for robustness by perturbing the SCRF parameters.
Hope this helps
Dominik
Hi Dominik,
Thanks so much for your response. I've gone ahead and convolved the simulated data for a subset of subjects with the SCRF as you mentioned. The model does fit nearly perfectly. However, I'm still unsure about the interpretation of the results. The result of US+ minus US- still has a negative mean. As does US+ minus CS-. Further, the mean and sem of both are nearly identical. Lastly, when looking at individual conditions (CS+r contrasts = 1, all other contrasts = 0), the means for the fixed response are 0 for CS+r and CS+u.
I am seeing a similar pattern with my real data, and can't help but wonder if maybe the delay in shock onset after the offset of my CS+ could be responsible (the CS+ is present for 2.5s and there is a 1s delay from the offset of the CS+ to the onset of the US, such that the shock arrives 3.5s after the onset of the CS+). In both real and simulated results, the flexible response results seem to make more sense than the fixed.
Any insights on this would be great.
Thank you,
Daria
Hi Daria
I understand. This is odd indeed. I agree that the temporal gap between CS and US could have an impact - it is possible that there is a response to the CS offset which is not properly modelled. If you want, I'd be happy to look at your new simulations and results, and it could be useful to look at the mean time course for the different conditions in your experiment. If you don't want to post this publicly, send me an email
Best
Dominik