HIGHLIGHTS
SUMMARY
Having found that the model can learn autonomously to solve arbitrary problems, the authors then aimed to identify which brain regions might show representations and activity that matched particular GOLSA model layers. To do this, the authors tested the GOLSA model with a Treasure Hunt task (Fig 2B and "Methods"), which was performed by both the GOLSA model and human subjects with fMRI. To analyze the fMRI and model data, the authors used model-based fMRI with representational similarity analysis (RSA)38 ("Methods"). The resulting fMRI RSA maps, one per GOLSA model layer, show . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.