first post-implementation discussion

[tzler]

Great talking today! A sketch of our conversation

there are some critical mistakes the authors made in their inclusion of the neutral stimuli

• basically, the equivalence doesn't work between neutral (0) and valenced (+ and -) stimuli
• i tried to be clear about the logic here in an sketch of my project proposal to my writeup, which also has some analysis descriptions
• this can be remedied with a simple fix:
  • add a button for neutral in acquisition ("press to proceed")
  • there are then 4 buttons (+, -, 0, novel) buttons and the analysis becomes symmetric
• this discussion is hidden somewhere in the bottom-middle of the white board

the authors seemed to have made a mistake in reporting what tones they tested

• graphs show different neutral intervals than they report in the methods
• this is a good thing to ask about when we email
• middle top of the whiteboard

a mood induction might be a really nice way to push around these "generalization" curves

• during the generalization stage, making subjects feel good should decrease the slope of their generalization curves (e.g. less error)
• top left of the whiteboard

this is certainly a memory task

• the authors don't show that people's perceptual thresholds actually change, just that their memories are effected by valence
• would be nice to test this directly by looking at whether just noticeable differences change around the acquisition stage
• bottom right in the whiteboard

we spoke about the different ways our data might align with theories of affect

• arousal only prediction is in the top right corner of the whiteboard
• valence specific hypothesis' prediction is just to the left of that

Next steps include

• talking with Mike about how true to the original our replication should be + funding
• emailing authors about details above + elsewhere -- basically, asking for the writeups and experiments so we don't have to guess
• building out a couple more details in the experiment

[mtheBee]

Additional next steps: tyler: Ask Mike if ok for us to add affect questions in between generalization trials now? Or later step? Maia: determine trait measures, affect post-trial questions