My failed “zombie” cognitive research, Part 1: imitating the classics
Hello ZombieLaw readers. Thank you for your readership. This post (and possibly the next few posts) is going to be quite a bit different. I have been working on various “zombie” research for years and failing so much. It’s finally time for me to quit on a portion of these efforts but I wanted to share the results with you all.
It’s been a very long road and some of the studies I tried are sort of interesting, but at this point none of them are particularly worth my time to continue. I’m not sure what that means for the future of this blog but the blog is a very different part of this project, and can still continue. However, in a larger sense it’s all sort of related so it will be interesting to see how my art develops.
None of this work I am going to report now has ever sparked the interest of my academic advisors. I’m done. I’m tired and burnt out. It’s not entirely the school’s fault, I was a difficult student, frequently bit the hand that fed me, my writing is disjointed and sometimes incomprehensible, I never liked formal academic style and clearly I still don’t get it. It’s ok. I think I’ll be happier without these zombies. But as I close off this time, let’s look back at some of my doomed efforts.
First, one of my early “zombie” cognition projects, I tried looking at the serial position effect. This is a classic of human memory that causes people to better remember words in the beginning and end of a list. Words in the middle of a list are more likely to be forgotten.
The word “zombie” seems a sticky word, and so I thought maybe it is more memorable than it should be. I put “zombie” in the middle of a list of words and the participants recall the words they remembered immediately at the end. I would need to do a lot more study to say for certain what is happening. Might need to try the word in other positions on the list and also test against other word options. In this study, the other words were selected from a database of word frequency in modern language but it’s possible that “zombie” had already become more frequent since the creation of that word frequency database. And also something is maybe going on with the word “reindeer” too. And some other irregularities. I tried a few iterations and these results happened:
After that study failed to spark my so-called advisor’s interest I moved next to exploring if there was something about the syllables themselves that made this word “zombie” special. I tried replicating the classic Bouba-Kiki experiments and replacing one of the words with “zombie” syllables. These results seemed to show that zombie could be perceived as either the round-shape or the pointy-shape depending on the contrast comparisons. Compared to Bouba or Kiki, “zombie” swings to the other.
In that table it says “right” and “left” but it means the first word of the comparison in the row heading. Notice a good repetition of the classic results for with Kiki being the sharp character and Bouba the rounded. The other versions show some of the effects of changing the -ie ending on zombie. Recall previous ZombieLaw posts made tangential reference to Bouba-Kiki research and possible syllable-effects: see “zombie linguistics scrabble dictionary” and “Noam Chomsky zombies! and why!“.
Who cares about shapes and word sounds? Not my so-called advisor. Round and sharp might be zombie false dichotomy and the effects are not clearly because of ambiguities in the sound-shape meaning (i.e. zom = rounded ; bie = sharp). Also zombie is a more familiar word than Bouba or Kiki. Who knows what it means? So I moved on…
I started asking people about their belief in zombies. Just a regular survey. I collected a bunch of participants on this survey via Amazon’s MTurk (which is how the previous studies were conducted too) and it seemed to show that the participants were more likely to think other people could be zombies if they also think they themselves might be a zombie. Duh? Too obvious, right? And the percentages of both were small so it’s not much of a finding.
There also seemed to be an age effect for affinity to zombies (based on likelihood of clicking an article with “zombie” in the headline title) but not a big difference in the ages. The average age of those with high zombie affinity was around 30 compared to about age 35 for those with lower affinities, but both groups with standard deviations larger than that difference. So again, not much of a finding. Also, that zombie-affinity variable (based on selecting headlines) also related to the question about whether other people might be zombies. Again this finding is somewhat obvious in that participants who prefer “zombie” headlines are also more inclined to be unsure about whether other people might be zombies. This doesn’t imply any actual effect of having these beliefs. Obviously an interest in zombies might increase the likelihood of considering the possibility that other people are zombies. It doesn’t mean these participants are processing any information differently. So again, not cognitive enough.
I did think I found a gender effect in the first sample of this survey that I collected. It was a somewhat large sample for cognitive research with over 100 participants, but unfortunately, these gender effects failed to reappear in the subsequent sample. Damn you, Type 1 error (or perhaps some unknown sampling bias – sometimes MTurk is an amazing service to collect survey opinions, but then sometimes the samples are so wildly different for no apparent reason. Anyway, none of my replicated effects were of particular interest to my so-called advisor, and so I moved on again…
Eventually, I hit upon a study idea that would become my Master’s thesis… unfortunately, not the doctorate I had hoped. Oh well, sorry Mom… but that’s a story for future posts… stay tuned… nom nom….