Sam Gershman

I love this project, which feels like it opens many doors.

Great opportunity here!!

@postlelab @JohnnyFoxe @EJNeuroscience @SfNtweets @JNeurophysiol @IBROjournals @JOCN_Journal Sorry, we don't... Maybe we should? But note that Open Mind isn't a society journal.

@JohnnyFoxe @EJNeuroscience @SfNtweets @JNeurophysiol @IBROjournals ooo ooo ooo! @JOCN_Journal @ Open Mind (@gershbrain does Open Mind had a twitter handle? not finding it...)

Precisely this! Support your society owned journals to keep the funds in the field! @EJNeuroscience @SfNtweets @JNeurophysiol @IBROjournals

@chrisXrodgers I directly believe you!

@gershbrain I think your "rhetorical flourish" hypothesis is reasonable. I notice in my own writing that I'm often tempted to use "direct" or "directly" as an almost meaningless intensifier ("Next, we directly tested this hypothesis by.."), similar to the way people sometimes use "very"

Papers that report neuronal recordings sometimes say that they provide "direct" evidence for neural encoding of something, in contrast to "indirect" methods such as non-invasive brain imaging. I don't really understand this. What makes one "direct" and the other "indirect"?

@PessoaBrain My point was not about to what extent methods measure neural activity directly (I think we all agree on that), but to what extent methods *test hypotheses about neural activity* directly, which depends a lot on what the hypotheses are.

@gershbrain fMRI BOLD is extremely indirect and depends in complex ways on blood oxygenation levels and even vessel size! While I would love for it to be a great index of neuronal activation, it's quite far from it. I use it because I study humans, not my preferred method!

@yyyxiang has a new paper with me and @kevin_dorst showing that you don't find a gambler's fallacy in probability judgments (by you do in point predictions): osf.io/preprints/psya… This dissociation suggests that the gambler's fallacy doesn't arise from probabilistic reasoning.

@VincentCostaPhd For sure. I actually shouldn't have been throwing blame around; my intention was to constructively call attention to an issue.

@gershbrain As someone who crosses that barrier regularly the fault isn’t just with systems neuroscientists. Plenty of neuroimagers can’t be bothered to consider neuroscience findings below the level of a voxel.

I think my point here has been misunderstood by most commentators. Obviously I agree that ephys is a more direct measure of spiking activity. Let me try to clarify...

@iampritishpatil I'm not coming down in favor of any particular hypothesis. I'm just making the point that different methods make sense for testing different hypotheses. Not every hypothesis about the role of neural activity in cognition can be answered using ephys.

@gershbrain That seems to push the onus on the null hypothesis. Is there a reason we think the non invasive signals always directly read out what the local neural activity is. Given the lack of mechanistic understanding or high correlation, it's not clear why we should assume this is true.

@martisamuser My point was that all of these things are measuring neural actvity (+other things). Which is the most appropriate measure depends on the specific hypothesis you're trying to test. The real hypothesis is almost never "neural vs non-neural".

My impression is that there is a cultural barrier where systems neuroscientists overlook the relevance of non-invasive brain imaging to test certain hypotheses, simply because it's not directly measuring spikes.

My point is that it's too simplistic to think that measuring spiking activity is always a more "direct" way of testing hypotheses about the relationship between neural activity and information processing.

@martisamuser My point wasn't about linearity but about causality. You need some neural activity (including LFPs) to generate BOLD signal, scalp EEG, etc, even if the relationship is complicated.

@martisamuser But all of that neural activity is ultimately spike-driven. So all of these measurements are reading out an aggregate effect of spiking activity, right?