Take the 2-minute tour ×
Cognitive Sciences Stack Exchange is a question and answer site for practitioners, researchers, and students in cognitive science, psychology, neuroscience, and psychiatry. It's 100% free, no registration required.

I'm fitting a psychometric function to a range of data. The majority of this data lends itself to a sigmoidal fit (i.e. participants can do the task), but some individuals are absolutely unable to do the task. I'm planning to compare the slopes obtained from different conditions, but I've hit a wall with the unable-to-do-the-task data.

enter image description here

Fitting a function to this data, the slope should be nearly flat, right? However, the data is really noisy and some weird fitting is occurring - I end up getting erroneously high slopes. I'm using pypsignifit, the parameters I'm using can be seen below. Any idea how to stop this happening?

num_of_block  = 7
num_of_trials = 20

stimulus_intensities=[3, 7, 13, 20, 27, 32, 39] # stimulus levels
percent_correct=[.38, .75, .6, .43, .7, .65, .43]     # percent correct sessions 1-3
num_observations     = [num_of_trials] * num_of_block      # observations per block
data= np.c_[stimulus_intensities, percent_correct, num_observations]
nafc = 1
constraints = ('unconstrained', 'unconstrained', 'unconstrained', 'Beta(2,20)' )
boot = psi.BootstrapInference ( data, core='ab', sigmoid='gauss', priors=constraints, nafc=nafc )
boot.sample(2000)
print 'pse', boot.getThres(0.5)
print 'slope', boot.getSlope()
print 'jnd', (boot.getThres(0.75)-boot.getThres(0.25))
share|improve this question
1  
The data look really like the participant performed on chance level. Actually, I would not try to fit them at all, because the fits cannot get better than your example. –  H.Muster Jun 3 '12 at 17:09
    
That's exactly what happened. It may be counterintuitive to try to fit the data, but I really want to be able to compare the slope of these participants with ones who performed better using something like a t-test. Hence the 'need' to fit. –  luser Jun 3 '12 at 18:09
    
Then I would try different psychometric functions (e.g, logistic, Weibull) until I find one that is fitted to the data as a straight line with a slope of almost zero. –  H.Muster Jun 3 '12 at 19:30
    
Makes sense, but the main issue I have is I'd have to arbitrarily choose when to use this alternative fit. The same participant appeared to perform at chance on another condition, but the cumulative gaussian fitted well and showed a not-immediately-apparent slope. –  luser Jun 3 '12 at 20:08
1  
I would not suggest to use different functions for individual fits, but use one functions for all, i.e., choose the function that gives the best overall solutions. Concerning the bad cases: did you try to seed the estimation routine with starting values nearer to 0.5 for the guessing rate and the lapsing rate? –  H.Muster Jun 4 '12 at 5:56

1 Answer 1

At the heart of the matter is the fact that 60% "yes" responses independent of the stimulus level (i.e., the problematic data) can arise from both an extremely sensitive subject (i.e., steep slope) with a moderate bias and a high lapse rate and an extremely insensitive subject (i.e., shallow slope) with a moderate bias and a low lapse rate. For your data the steep slope/high lapse rate fit is slightly better than the shallow slope/low lapse rate when your prior on the lapse rate is based on a Beta distribution. My guess is if you used a uniform prior on the lapse rate, and possibly on the guessing rate, this will result in the shallow slope fit being better. I would try something like "Uniform(0,0.1)".

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.