Goeienavond allen!
Hopelijk zijn jullie goed thuis gekomen!
De co-auteur van mevrouw Sharot heeft veel vragen beantwoord. Ik heb nu geen puf om het inhoudelijk te bekijken, maar ik stuur het vast door.
Goed weekend!
Hartelijke groet, Lisette
________________________________ Van: Christoph Korn [christoph.w.korn@gmail.com] Verzonden: donderdag 29 augustus 2013 18:08 Aan: Harting, Lisette CC: Tali Sharot Onderwerp: Re: Question about research paper
Dear Lisette,
Thank you very much for your interest in our study. Please see below for my answers:
2013/8/28 Harting, Lisette <L.P.Harting@uva.nlmailto:L.P.Harting@uva.nl> Dear Tali Sharot and Christoph Korn,
Thank you very much for your swift reply! We're looking forward to Chris' answer!
Best wishes, Lisette Harting ________________________________ Van: Tali Sharot [talisharot@googlemail.commailto:talisharot@googlemail.com] Verzonden: woensdag 28 augustus 2013 23:03 Aan: Harting, Lisette CC: Christoph Korn Onderwerp: Re: Question about research paper
Hello, I am cc-ing my co-author Chris who should be able to help you out. Best, Tali
On Wed, Aug 28, 2013 at 11:32 AM, Harting, Lisette <L.P.Harting@uva.nlmailto:L.P.Harting@uva.nl> wrote: Dear Tali Sharot,
My colleagues (teachers of the first year course 'Academic Basic Skills' for psychobiology students at the University of Amsterdam) and I have been discussing your research paper 'Selectively altering belief formation in the human brain'. The reason is that our first year psychobiology students will next week start their first year by writing a 'research description' (1100 word summary) based on this paper. During the preparation discussion, we were left with some questions, that your 2011-paper could not shine enough light onto for us to fully understand all of it. Would you maybe be so kind to help us out with the following questions?
-Question 1- Method / timeline We were a bit puzzled by what you exactly mean by sessions, trials and blocks, while assembling a time line. Is the following correct? - A subject starts with block 1. This block can contain either only 'happen' questions or only 'not happen' questions. - In this block, there are 20 trials. - Is the memory test at the end of all other tests, or integrated in the blocks? - Are there any (long term) breaks between any parts of the total experiment? If so, when and how long?
- Each trial consists of two sessions that occur consecutively: We discussed, read, discussed and read again and more, but we couldn't arrive at a compromise: - Which of the following is true? - Option 1: Block 1 Subject A: session A1, session A2, (memory test) Subject B: session B1, session B2, (memory test) Subject C: session C1, session C2, (memory test) ... (20 subjects)
Block 1 Subject G: session G1, session G2, (memory test) Subject H: session H1, session H2, (memory test) Subject I: session I1, session I2, (memory test) ... (20 subjects)
--> This is how I would describe it:
Subject A: - Session 1 - Happen block (20 trials i.e. 20 events) - Session 2 - Happen block (20 trials i.e. same 20 events as in session 1 - happen) - Session 1 - Not_happen block (20 trials i.e. 20 events) - Session 2 - Not_happen block (20 trials i.e. same 20 events as in session 1 - not_happen) - memory test - ratings (familarity, etc.)
Subject B: - Session 1 - Not_happen block (20 trials i.e. 20 events) - Session 2 - Not_happen block (20 trials i.e. same 20 events as in session 1 - not_happen) - Session 1 - Happen block (20 trials i.e. 20 events) - Session 2 - Happen block (20 trials i.e. same 20 events as in session 1 - happen) - memory test - ratings (familarity, etc.)
Subject C like A Subject D like B etc.
There were no longer pauses between the different parts (except for short explanations). There were different lists of each 20 events (which were counterbalanced for order).
- Option 2: Block 1 Session 1: Subject A, subject B, subject C, ... (20 subjects) Session 2: Subject A, subject B, subject C, ... (20 subjects) Memory test: Subject A, subject B, subject C, ... (20 subjects)
Block 2 Session 1: Subject G, subject H, subject I, ... (20 subjects) Session 2: Subject G, subject H, subject I, ... (20 subjects) Memory test: Subject G, subject H, subject I, ... (20 subjects)
There may be other options, I suppose.
-Question 2- What is the percentage of received good news for the subjects shown in figure 2A (or on average per group). Is it so that people are quite positive and therefore get relatively more bad news? Is there a difference in this baseline-positivity per group and if so, could it have influenced the results? Is there asymmetry in the size of the errors if you compare the 'bad news errors' with the 'good news errors' (this could influence the good news bad news effect size)? --> On average, people had roughly as many good as bad trials. Yes, you are absolutely right; "baseline-positivity" could theoretically have influenced the results. What you call "baseline-positivity" corresponds to "initial estimates" in the paper (see last paragraph of results section and table 1). We made sure that there was no difference between groups. See also our other papers using the same task, in which the same issue is mentioned (Current Biology, Nature Neuroscience, and Psychological Medicine).
-Question 3- Could you maybe provide us with the list of questions, so we can show our students more examples? --> See attached; A list with all items that we used and their respective percentages.
Let me know if things have remained unclear or if additional questions come up during the discussions with your students.
Best, Chris
Thank you very much for your time and help! If you need more information, please let me know!
Best wishes,
Lisette Harting ---- Teacher Academic Basic Skills for Psychobiology University of Amsterdam -- -------------------------------------------------------------------------------------------------------------------------- Affective Brain Lab. Cognitive Perceptual & Brain Science. University College London. http://www.ucl.ac.uk/affective-brain/