Little by little
Despite very warm weather, some flies did fly, even though the learning performance of the control flies was really poor. At least for now, it looks like all stocks are learning and that rut and rsh flies learn at least equally well as the Berlin flies. I’ve also managed to fix the positive preference problem:
Kicked out Canton S
Since the Canton S strain I used wasn’t a perfect genetic background strain anyway and didn’t fly properly, I kicked it out and replaced it with the wild type Berlin data I had collected just prior to the rut and rsh mutants. Now I can collect data for two projects in one go: I check the rut/rsh learning mutants if they still behave the way they should and with the wtb control strain and continue collecting data to evaluate their optomotor responses after training. So now the learning scores for the three strains look like this:
It still doesn’t look like rut is better than wtb, so I’m still skeptical that the strain is really what it should be. The rsh data also don’t look very promising, but I still need to get more flies with a negative preference before the training.
Suspicion and progress!
Suspicion: rut may not be rut
Carsten Duch is identifying the muscles involved in operant self-learning:
It’s a start
The first few rutabaga and radish mutants have been measured and so far it looks good: standard duration yaw torque learning – the situation all three groups should learn.
Optomotor asymmetry after torque learning
Only a few more flies compared to the first week, but the difference between optomotor responses after punishing left or right torque, respectively, is starting to show:
These data suggest I should fill the groups up to a sample size beyond 20 if I want to use these data somewhere…
Optomotor effects after torque learning?
After training 14 flies it looks as if the new setup is working as it should:
Looking at the optomotor traces more closely now that (a) we know that motor neurons are the only site of plasticity and (b) we routinely record optomotor behavior after training, something very interesting appeared when comparing flies that have been trained on one turning direction with flies trained on the other side:
Flies trained to avoid right turning show reduced optomotor response to the right and vice versa! The weaker effect for ‘left’ may be due to the weaker learning in this group?
Anyway, I have never seen this and find it quite exciting!
Machine works, flies are not learning
The old software is installed with the new hardware and I’ll be testing the new software in parallel to experiments with the old software. Technically, everything seems to be working now, but the first six flies I tested did not show learning after training:
However, avoidance looks fairly ok, only the last training period gives away that something still isn’t quite the way it should be:
So I’ll go ahead and train a new batch of flies this week. I’ll focus on training/avoidance and make sure it is perfect, before I shorten the training.
NWG Poster on MBON-02 regulating habit formation
If you missed it at the Göttingen NWG meeting, you can download our poster from here:
Numerosity in Buridan’s
Last week, high-school student Marie Grünleitner tested wild type and blind flies in a modified Buridan’s paradigm. Instead of just two stripes, she put four stripes onto the arena walls: one stripe on one side and three stripes on the other. This experiment was inspired by a poster at FENS22 which is summarized in this preprint. She tested 15 flies in each group. She found that wild type flies prefer the threes stripes (up in the figures) while blind flies stay mostly anywhere at the edge of the platform.
Wild type flies:
NorpA blind flies:
I will use the other software from the authors in Paris to see if it runs on our data and what results these evaluations will yield.