I am an experienced interdisciplinary researcher and have focused on using computational methods to derive insights into biological systems. My academic research took me from collecting insects in tropical rainforests to imaging them in synchrotrons. I have become progressively more involved with the Effective Altruism community over several years and I am now aiming to apply my expertise in areas that more directly benefit society. To that end, I have recently redirected my research towards exploring novel technical countermeasures against viral pandemics.
gavintaylor
The illusion of science in comparative cognition
The Intellectual and Moral Decline in Academic Research
Nice post Daniela.
You might be interested in the concept of the Devonian Toolkit raised a few years ago. It is mostly concerned with sensorimotor control in flight and suggests that because insect brains have remained relatively similar since the Devonian, and most flight control behaviours found in Drosophila are also found in other flying insects (and there are often similar behaviours for walking control), flying insects are likely to share a similar set of basic behavioural modules for flight control (the Devonian toolkit). This is hard to test but seems reasonable. I’d suggest that it is also possible that phylogenetically distant insects with a common ancestor may share similar capacities in the indicators this study uses for consciousness (like learning and memory traits). Thus, although insecta is a large taxa positive findings for some indicators could be quite generalizable.
I also have a comment about positive publication bias, particularly in behavioral experiments. Although such a bias certainly exists, and lack of evidence against a trait is not the same as evidence against a trait, evidence against a trait also has quite a high likelihood of being a false negative in a behavioral experiment. The reason for this is that even if an invertebrate is capable of displaying some behavioral trait, it can be quite hard design the correct experimental paradigm to encourage them to display it. In general I’d be put more faith in negative results reported for reflexive behaviours; anything involving a training task can become really difficult if the animal isn’t motivated to participate. For example, I’ve had colleagues working with polychaetes and toads (ok, not an invertebrate) who both struggled just to get these animals to respond reflexively to big obvious visual stimuli, let alone train them to do perform a discrimination task.
Nice article Jason. I should start by saying that as a (mostly former) visual neuroscientist, I think that you’ve done quite a good job summarizing the science available in this series of posts, but particularly in these last two posts about time. I have a few comments that I’d like to add.
Before artificial light sources, there weren’t a lot of blinking lights in nature. So although visual processing speed is often measured as CFF, most animals didn’t really evolve to see flickering lights. In fact, I recall that my PhD supervisor Srinivasan did a study where he tried to behaviorally test honeybee CFF—he had a very hard time training them to go to flickering lights (study 1), but had much more success training them to go to spinning disks (study 2). In fact, the CFF of honeybees is generally accepted to be around 200 Hz, off the charts! That said, in an innate preference study on honeybees that I was peripherally involved with, we found honeybee had preferences for different frequencies of flickering stimuli, so they certainly can perceive and act on this type of visual information (study 3).
Even though CFF has been quite widely measured, if you wanted to do a comprehensive review of visual processing speed in different taxa then it would also be worth looking at other measures, such as visual integration time. This is often measured electrophysiologically (perhaps more commonly than CFF), and I expect that integration time will be at tightly correlated with CFF and as they are causally related, one can probably be approximately calculated from the other (I say approximately because neural nonlinearities may add some variance, in the case of a video system it can be done exactly). For instance, this study on sweat bees carefully characterized their visual integration time at different times of day and different light conditions but doesn’t mention CFF.
Finally, I think some simple behavioural experiments could shed a lot of light on how we expect metrics around sensory (in this case visual) processing speeds to be related to the subjective experience of time. For instance, the time taken to make a choice between options is often much longer than the sensory processing time (e.g. 10+ seconds for bumblebees, which I expect have CFF above 100 Hz), and probably reflects something more like the speed of a conscious process than the sensory processing speed alone does. A rough idea for an experiment is to take two closely related and putatively similar species where one had double the CFF of the other, measure the decision time of each on a choice-task to select flicker or motion at 25%, 50% and 100% of their CFF. So if species one has CFF at 80 Hz, test it on 20, 40 and 80 Hz, and if species two has CFF 40 Hz, test it on 10, 20 and 40 Hz. A difference in the decisions speed curve across each animals frequency range would be quite suggestive of a difference in the speed of decision making that was independent of the speed of stimulus perception. The experiment could also be done on the same animal in two conditions where its CFF differed, such as in a light- or dark-adapted state. For completeness, the choice-task could be compared to response times in a classical conditioning assay, which seems more reflexive, and I’d expect differences in speeds here correlate more tightly to differences in CFF. The results of such experiments seem like they could inform your credences on the possibility and magnitude of subjective time differences between species.
Working together to examine why BAME populations in developed countries are severely affected by COVID-19
This article on doing systematic reviews well might also be of interest if you want to refine your process to make a publishable review. It’s written by environmental researchers, but I think the ideas should be fairly general (i.e. they mention Cochrane for medical reviews).
I’d also recommend having a loot at Iris.ai. It is a bit similar to ConnectedPapers but works off a concept map (I think) rather than than a citation map, so it can discover semantic linkages between your paper of interest and others that aren’t directly connected through reference links. I’ve just started looking at it this week and have been quite impressed with the papers it suggested.
The idea of doing deliberate practice on research skills is great. I agree that learning to do good research is difficult and poor feedback mechanisms certainly don’t help. Which other skills are you aiming to practice?
This seems like a thorough consideration of the interaction of BCIs with the risk of totalitarianism. I was also prompted to think a bit about BCIs as a GCR risk factor recently and had started compiling some references, but I haven’t yet refined my views as much as this.
One comment I have is that risk described here seems to rely not just on the development of any type of BCI but on a specific kind, namely, relatively cheap consumer BCIs that can nonetheless provide a high-fidelity bidirectional neural interface. It seems likely that this type of BCI would need to be invasive, but it’s not obvious to me that invasive BCI technology will inevitably progress in that direction. Musk hint’s that Neuralink’s goals are mass-market, but I expect that regulatory efforts could limit invasive BCI technology to medical use cases, and likewise, any military development of invasive BCI seems likely to lead to equipment that is too expensive for mass adoption (although it could provide the starting point for commercialization). Although DARPA’s Next-Generation Nonsurgical Neurotechnology (N3) program does have the goal of developing high-fidelity non- or minimally-invasive BCIs; my intuition is at that they will not achieve their goal of reading from one million and writing to 100,000 neurons non-invasively, but I’m not sure about the potential of the minimally-invasive path. So one theoretical consideration is what percentage of a population needs to be thought policed to retain effective authoritarian control, which would then indicate how commercialized BCI technology would need to be before it could become a risk factor.
In my view, a reasonable way to steer BCIs development away from posing a risk-factor for totalitarianism would be to encourage the development of high-fidelity non-invasive and read-focused consumer BCI. While non-invasive devices are intrinsically more limited than invasive ones, if consumers can still be satisfied by their performance then it will reduce the demand to develop invasive technology. Facebook and Kernel already look like they are moving towards non-invasive technology. One company that I think is generally overlooked is CTRL-Labs (now owned by Facebook), who are developing an armband that acquires high-fidelity measurements from motor neurons—although this is a peripheral nervous system recording, users can apparently repurpose motor neurons for different tasks and even learn to control the activity of individual neurons (see this promotional video). As an aside, if anybody is interested in working on non-invasive BCI hardware, I have a project proposal for developing a device for acquiring high-fidelity and non-invasive central nervous system activity measurements that I’m no longer planning to pursue but am able to share.
The idea of BCIs that punish dissenting thoughts being used to condition people away from even thinking about dissent may have a potential loophole, in that such conditioning could lead people to avoid thinking such thoughts or it could simply lead them to think such thoughts in ways that aren’t punished. I expect that human brains have sufficient plasticity to be able to accomplish this under some circumstances and while the punishment controller could also adapt what it punishes to try and catch such evasive thoughts, it may not always have an advantage and I don’t think BCI thought policing could be assumed to be 100% effective. More broadly, differences in both intra- or inter-person thought patterns could determine how effective BCI is for thought policing. If a BCI monitoring algorithm can be developed using a small pool of subjects and then applied en masse, that seems much risky than if the monitoring algorithm needs to be adapted to each individual and possibly updated over time (though there would be scope for automating updating). I expect that Neuralinks future work will indicate how ‘portable’ neural decoding and encoding algorithms are between individuals.
I have a fun anecdotal example of neural activity diversity: when I was doing my PhD at the Queensland Brain Institute I did a pilot experiment for an fMRI study on visual navigation for a colleague’s experiment. Afterwards, he said that my neural responses were quite different from those of the other pilot participant (we both did the navigation task well). He completed and published the study and ask the other pilot participant to join other fMRI experiments he ran, but never asked me to participate again. I’ve wondered if I was the one who ended up having the weird neural response compared to the rest of the participants in that study… (although my structural MRI scans are normal, so it’s not like I have a completely wacky brain!)
The BCI risk scenario I’ve considered is whether BCIs could provide a disruptive improvement in a user’s computer-interface speed or another cognitive domain. DARPA’s Neurotechnology for Intelligence Analysts (NIA) program showed that an x10 increase in image analysis speed with no loss of accuracy, just using EEG (see here for a good summary of DARPAs BCI programs until 2015). It seems reasonable that somewhat larger speed improvements could be attained using invasive BCI, and this speed improvement would probably generalize to other, more complicated tasks. When advanced BCIs is limited to early adopters, could such cognitive advantages facilitate the risky development in AI or bioweapons by small teams, or give operational advantages to intelligence agencies or militaries? (happy to discuss or share my notes on this with anybody who is interested in looking into this aspect further)
Are there any areas covered by the fund’s scope where you’d like to receive more applications?
Something that I think EAs may be undervaluing is scientific research done with the specific aim of identifying new technologies for mitigating global catastrophic or existential risks, particularly where these have interdisciplinary origins.
A good example of this is geoengineering (the merger of climate/environmental science and engineering) which has developed strategies that could allow for mitigating the effects of worst-case climate change scenarios. In contrast, the research being undertaken to mitigate worst-case pandemics seem to focus on developing biomedical interventions (biomedicine started as an interdisciplinary field, although it is now very well established as its own discipline). As an interdisciplinary scientist, I think there is likely to be further scope for identifying promising interventions from the existing literature, conducting initial analysis and modelling to demonstrate these could be feasible responses to GCRs, and then engaging in field-building activities to encourage further scientific research along those paths. The reason I suggest focusing on interdisciplinary areas is that merging two fields often results in unexpected breakthroughs (even to researchers from the two disciplines involved in the merger) and many ‘low-hanging’ discoveries that can be investigated relatively easily. However, such a workflow seems uncommon both in academia (which doesn’t strongly incentivise interdisciplinary work or explicitly considering applications during early-stage research) and EA (which [with the exception of AI Safety] seems to focus on finding and promoting promising research after it has already been initiated by mainstream researchers).
Still, this isn’t really a career option as much as it is a strategy for doing leveraged research which seems like it would be better done at an impact focused organisation than at a University. I’m personally planning to use this strategy and will attempt to identify and then model the feasibility of possible antiviral interventions as the intersection of physics and virology (although I haven’t yet thought much about how to effectively promote any promising results).
Good point. Unfortunately the Economist article referenced for this number is pay-walled for me and I am not sure if it indicates the total number of clinical trial participants during that time.
Your comment got me interested so I did some quick googling. In the US in 2009 there were 10,974 registered trials with 2.8 Million participants, and in the EU the median number of patients studied for a drug to be approved was 1,708 (during the same time window). I couldn’t quickly find the average length of a clinical trial.
I expect 80,000 patients would be at most 1% of population of total clinical trial participants during that 10 year window, so this claim might be a bit over-emphasised (although it does seem striking at first read).
This might be the first example I’ve seen of an Open Inverse Grant Proposal. Good luck!
I’m interested in seeing a second post on impact purchases and would personally consider selling impact in the future. I have a few general comments about this:
Impact purchases seem similar to value-based fees that are sometimes used in commercial consulting (instead of time- or project-based fees) and may be able to provide a complementary perspective. Although in business the ‘impact’ would usually be something easy to track (like additional revenue) and the return the consultant gets (like percentage of revenue up to a capped value) would be agreed on in advance. I wonder if a similar pre-arrangement for impact purchase could work for EA projects that have quantifiable impact outcomes, such as through a funder agreeing to pay some amount per intervention distributed, student educated, etc. Of course, the tracked outcome should reflect the funders true goals to prevent gaming the metric.
It seems like impact purchases would be particularly helpful for people coming into the EA community who don’t yet have good EA references/prestige/track-record but are confident they can complete an impactful project, or who want to work on unorthodox ideas that the community doesn’t have the expertise to evaluate. If they try something out and it works then they can get funds to continue and preliminary results for a grant, if not, it’s feedback to go more mainstream. For this dynamic to work people should probably be advised to plan relatively short projects (say a up too few months), otherwise they could spend a lot of time on something nobody values.
This could be a particularly interesting time to trial impact purchases used in conjunction with government UBI (if that ends up being fully brought in anywhere). UBI then removes the barrier of requiring a secure salary before taking on a project.
From my experience applying to a handful of early-career academic grants and a few EA grants, I agree that almost none provide any/useful feedback (beyond accepted or declined), either for the initial application or for progress or completion reports. However, worse than having no feedback is that I once heard from an European Research Council (ERC) grant reviewer that their review committees are required to provided feedback on rejected applications, but also instructed to make sure the feedback is vague and obfuscated so the applicant will have no grounds to ask for an appeal, which means the applicant gets feedback the reviewers know won’t be useful for improving their project… Why do they bother???
With regards to implementation. I think one point to consider is the demand from impacters relative to funds of purchasers. At least in academia, funding is constrained and grant success rates are often <20%, and so grantees know that it is unlikely they’ll get a grant to do their project (academic granters often say they turn away a lot of great projects they want to fund). If impact purchasers were similarly funding constrained relative to the number of good projects, I think the whole scheme would be less appealing as then even if I complete a great project, getting its impact bought would still involve a bit/lot of luck.
These posts about impact prizes and altruistic equity may also be of interest to consider.
Very interesting post! I have worked in life science up to the postdoc level and think that is generally a reasonable summary of how life sciences research works (disclosure, Guzey interviewed me for this study).
One question is I have is how generalizable is this description geographically and across Universities? Based on the Universities/funders referenced I’d assume your thinking about Tier 1 Research Universities in the US. But did the demographics of your interviewee demographics suggest this could be situation more broadly?
A few other comments to e on some of the points:
Role of PIs
Agreed that senior PIs with large labs tend not to do very much bench work themselves. However, they aren’t solely managing and writing grants—I think one of the most important things PIs do is knowledge synthesis through writing literature reviews. I haven’t really met any postdocs that have the depth and breadth of knowledge of their lab head, which allows the later to both provide a high-level summary of their fields in reviews and also propose new ways forward in their grants.
A counterpoint I’ve come across is in mixed labs runs by a PI with a computational background who has postdocs and PhDs doing lab work while he works on using their biological results for computational modelling. From my perspective, these types of labs seem to function quite well as the PI usually relies on people coming into the lab to be well trained in the biological assays they’ll use, but then teaches them computational techniques that end up using themselves by the end of their project.Peer review
One of the big drawbacks of peer review is the hugely variable quality of reviews that are provided. As an example simply in terms of the level of detail provided, I have had comments of one paragraph and three pages for the same article.
I think a key reason for this is there isn’t really any standardized format or expectations for reviews nor is there much training or feedback for reviewers. One thought I’ve had is that paying peer-reviewers would allow journals to both enforce review consistency and quality—although publishers have such large profit margins that it this could be feasible, they have no incentive to do so as scientists accept the status quo. In the absence of paid peer-review, I think that disclosing reviewer names and comments helps prevent ‘niche guarding’ and encourage reviewers to provide a useful and honest review (eLife does this currently, not sure if any other journals do so).Permanent researchers
Agreed that letting postdocs move into staff scientist/researcher positions would be helpful—this has been discussed a bit in the Nature and Sciences career sections over the last few years (such as here). I’ve usually heard from postdocs who moved into staff scientist or lab/facility manager positions that they wanted to stop relying on grants for their employees and to get some job stability. But some then later regretted the move after finding the positions didn’t have many options for career advancement relative the professor track. The staff scientists role is a relatively new academic position (although it has been around for a long time in government and private research labs) that doesn’t yet have a lot of consistency between Universities—it would probably help to have more discussion and even formalize the roles expectations before a lot of people move into it.Solo founders
This is an interesting observation and I hadn’t thought about the individual lab head model in this way. I’d actually like to take this a step further and say that academia has a habit of breaking up good pairs of biologists. How so? In a few cases, I’ve seen two senior postdocs or a postdoc and junior PI (so essentially two researchers quite closely matched their level of experience and with complementary skills) work really well together and produce outstanding results over a few years, which will usually lead to one of the duo getting a permanent position. The other may be able to continue on as a postdoc for a while, but as their research speciality will overlap heavily with their colleague’s field and it’s unlikely that the hiring/promoting institution will open another position in a similar area for a few years, the postdoc will probably have to move elsewhere to continue their career. Although the two may continue to collaborate, the second person to be hired often starts working on different topics to show their intellectual independence (although the new topics may be less impactful than what they were working on as a pair). I only know of a few cases where duos separated in this way and I haven’t really followed their outcomes, but I’d assume that the productivity of both researchers declined afterwards. Allowing one to move into a staff researcher position would help in this respect.Big labs vs. small labs
Another option is a cluster of small labs working on a similar theme (I was in one in Lund that worked on Vision, another in the department worked on Pheromones). This seems to be more common in Northern Europe where high salaries tend to limit the group sizes that are possible (often PI, 1-2 postdocs, 1-2 PhDs). Clusters seemed to have the benefits noted for larger labs, but meant there were a lot of PIs around to mentor students, and also allowed the cost of lab facilities and support staff to be shared.Research niches
Territorial PIs seem quite common, and as noted, the publication/grant review process allows them to be quite effective at delaying/blocking and even stealing ideas that encroach on their topic. A link was recently posted here to an economics paper taht even suggested new talent entering a field after the death of a gatekeeping PI could speed up research progress. If it seems that a gatekeeping PI is holding back research in an important field, I think that a confrontational grantmaking strategy could be used—whereby a grant agency offers to fund research on the topic but explicitly excludes the PI and his existing collaborators from applying and reviewing proposals.Differing risk-aversion between PIs and students
Although a PI may seem risk-loving, he benefits from being able to diversify his risk across all of his students and may only need one to get a great result to keep the funding coming. He’s unlikely to get all of his students working together on one hard problem, just like a student can’t spend all his time on a high-risk problem.
I tend to think that developing the ability to judge a project’s risk is an important skill during a PhD, and a good supervisor should be able to make sure student has at least one ‘safe’ project that they can write up. Realistically it is possible to recover from a PhD where nothing worked well during a postdoc, but it is a setback (particularly in applying for ECR fellowships).
I feel that postdocs are possibly where the highest risk projects get taken on at the individual level, both because they have the experience to pick an ambitious but achievable goal, and also because they want to publish something great to have a good chance at a faculty position.
Ok, finally got around to writing about navigation. A few comments I have about this:
-I agree that navigating known paths/areas is a fairly simple skill. However, if the animal increases the speed with which it traverses such areas it is usually taken as an indicator of becoming familiar with the route. If an animal always moves at a constant speed in a known environment, it may be an indicator that it is constantly in exploring without ever learning.
-The examples presented for navigating unknown areas in the Sentience Table are a bit less clear for me in terms of whether they reflect navigational learning or contextual conditioning. Mazes (as they are generally presented to humans) do seem a reasonable indicator of learning to navigate an unknown area, however, the way they are often used in insect studies means that they primarily test conditioning rather than navigational ability. For instance, the methods used to teach bees to navigate a maze in Zhang et al 2000, were:
Bees were trained to come to a feeder placed initially just outside the entrance to the maze. After they were marked, the feeder was moved slowly step by step through the maze, remaining for ∼1 h in each decision chamber.
As such, it seems to more of an indicator they learnt a series of choices they had to take quite slowly. Likewise, Zhang et al 1996 show bees learning symbolic cues to solve mazes (such as turn right if the wall is green) seem to be more of an indicator of rule learning.
The Drosophila heat aversion paradigm developed by Ofstad et al is quite similar to the Morris water maze, and although this paradigm is a good test of visual-spatial memory (when the animal then quickly changes its position to the new cool point based on movement in the visual panorama), reaching the safe point should be solvable by a Type 1 Braitenberg vehicle (which does not seem to be intelligent).
The examples of maze learning in cockroaches are perhaps a bit more like what humans generally associate with maze learning—I looked back through the references from Webb and Wystrach 2016 and found the original paper on maze learning in cockroaches, where roaches navigated an actual hot metal maze to find a cooler safe point, and it seems their speed and accuracy increased over time.
Perhaps an issue is that maze learning is difficult to motivate insects to do in the same way that vertebrates do. For instance, I think it would be very hard to train a bee to enter a maze and search it for food—placing it (or the entire hive) at the centre seeing if they navigate out might be a better analogy (but I suspect this may just end up with them getting stuck in the corners). That said, I think it is fairly clear that central place foragers navigate unfamiliar territories, it’s just that I don’t find most uses of mazes to be particularly relevant. The fact that a honeybee hive can be moved to a forest and the bees will quickly forage on available flowers seems a good indication of their ability to navigate unknown areas, but I don’t know of anybody who has really tried to quantify this, it’s just taken as a given.
-When discussing spatial memory, it’s important to consider the distinction of traversing routes vs. having a map like memory. Traversing route (or things like traplining) implies that a set path can be learnt (indicated by landmarks or odometry) but not necessarily that different paths can be linked. However, map memory is taken to imply that routes are placed on a topographic representation in its memory and that an animal can then use this map to link points on known routes with a novel shortcut (that isn’t based on shared landmarks visible between the routes). This is quite controversial and hard to motivate insects to do reliably (as bees and ants tend to try to go to and from their nest on specific routes, but don’t usually jump between routes). I would place this higher than detouring in terms of navigational ability. Actually, I was surprised to see detouring as a navigational ability as I’d never thought about it much. However, I agree that ant work indicates detouring shows a degree of navigational flexibility between direct route following and map navigation. Unfortunately it’s probably quite hard to test detouring reliably in flying insects without building large 3D constructs, although some virtual reality work may have done this.I’ve enjoyed looking through the criteria and evidence you’ve used in putting together the Invertebrate Sentinance Table, particularly in that its led me to think place my knowledge of invertebrate neuroscience in a consciousness framework. Feel free to get in touch if you’d like my opinion on any of your further work on this.
At the start of Chapter 6 in the precipice, Ord writes:
To do so, we need to quantify the risks. People are often reluctant to put numbers on catastrophic risks, preferring qualitative language, such as “improbable” or “highly unlikely.” But this brings serious problems that prevent clear communication and understanding. Most importantly, these phrases are extremely ambiguous, triggering different impressions in different readers. For instance, “highly unlikely” is interpreted by some as one in four, but by others as one in 50. So much of one’s work in accurately assessing the size of each risk is thus immediately wasted. Furthermore, the meanings of these phrases shift with the stakes: “highly unlikely” suggests “small enough that we can set it aside,” rather than neutrally referring to a level of probability. This causes problems when talking about high-stakes risks, where even small probabilities can be very important. And finally, numbers are indispensable if we are to reason clearly about the comparative sizes of different risks, or classes of risks.
This made me recall hearing about Matsés, a language spoken by an indigenous tribe in the Peruvian Amazon, that has the (apparently) unusual feature of using verb conjugations to indicate the certainty of information being provided in a sentence. From an article on Nautilus:
In Nuevo San Juan, Peru, the Matsés people speak with what seems to be great care, making sure that every single piece of information they communicate is true as far as they know at the time of speaking. Each uttered sentence follows a different verb form depending on how you know the information you are imparting, and when you last knew it to be true.
...
The language has a huge array of specific terms for information such as facts that have been inferred in the recent and distant past, conjectures about different points in the past, and information that is being recounted as a memory. Linguist David Fleck, at Rice University, wrote his doctoral thesis on the grammar of Matsés. He says that what distinguishes Matsés from other languages that require speakers to give evidence for what they are saying is that Matsés has one set of verb endings for the source of the knowledge and another, separate way of conveying how true, or valid the information is, and how certain they are about it. Interestingly, there is no way of denoting that a piece of information is hearsay, myth, or history. Instead, speakers impart this kind of information as a quote, or else as being information that was inferred within the recent past.
I doubt the Matsés spend much time talking about existential risk, but their language could provide an interesting example of how to more effectively convey aspects of certainty, probability and evidence in natural language.
- Matsés—Are languages providing epistemic certainty of statements not of the interest of the EA community? by 8 Jun 2021 19:25 UTC; 15 points) (
- Matsés—Are languages providing epistemic certainty of statements not of the interest of the EA community? by 8 Jun 2021 19:25 UTC; 15 points) (
Thanks for the comments Peter and Michael. I’m not very familiar with the mirror test in general so I can’t comment with confidence about how well this compares to the results with other species. But after having looked back at Table 2 in the paper reporting the mirror test results I’d argue the results aren’t so clear cut—one ant never cleaned itself, whereas the other ants cleaned themselves between 1 and 9 times over the six minute trial (also, the behaviour never occured in juvenile ants). I don’t think this indicates that ants are smarter than chimps, another explanation is simply that, assuming cleaning occurs because the ants visually recognised the paint spot from their reflection, that this triggered a reflexive grooming behaviour. Chimps probably have more complex motivations—if they see the paint spot some may want to remove it, but others might not be in the mood for cleaning or could enjoy having it there. If the difference is then between reflexive ant cleaning vs. voluntary chimp cleaning you could then go onto discuss the relevance of each type of behaviour for demonstrating self-recognition, but I don’t think we are there yet.
Admittedly, this study was not performed rigorously. The review Michael links to (Gallup 2018) presents some important criticism. The ants also could have perceived the paint through mechanosensory hairs on their head—although the difference in grooming between brown and blue paint seems to suggest this wasn’t the case. Another point is that this study doesn’t really define what was considered cleaning behaviour, and it may be that the threshold for this was lower than that used in by chimp researchers, raising the likelihood of false positives for the ants. In addition to clear monitoring of the ant grooming pre-, during, and post- mirror exposure, it would also be useful to do a recovery experiment whereby the blue paint was covered with brown paint—if this prevented later grooming behaviour it would strongly suggest that the grooming was related to the ants visual perception of its reflection.
It’s also a fair point that both Caemmarts do not perform very rigorous research studies and generally publish them in lowly ranked to predatory journals. Publishing in low ranked journals isn’t a crime in itself, although the peer review usually isn’t very rigorous and I wouldn’t be surprised if these authors simply keep submitting the same manuscript to a journal until one will take it without revisions (this is poor scholarship but not uncommon). However, it does present a bit of problem if the first report on a controversial topic (invertebrate self-recognition) comes out like this—it probably didn’t get much attention from other ant cognition researchers (there might not be many anyway), and a simple replication study can’t be published in highly ranked journal (if the results held up and the study was done well I think this would have made it into mid to top ranked generalist journal).
In reality, this is a pretty simple experiment to replicate with ants. I know a few ant navigation researchers in Europe who could easily supervise a student to replicate it over summer (the Myrmica genus is quite common). I can put you in contact if this is a useful point for RP to confirm for the invertebrate sentience project? I’d also wager $50 that this does replicate, insofar that the ants groom paint spots based on their reflection but besides any interpretation of self-awareness based on this result.
Is there a Price for a Covid-19 Vaccine?
I agree that it’s an extreme stance and probably overly-general (although the specificity to public health and biomedical research is noted in the article).
Still, my feeling is that this is closer to the truth than we’d want. For instance, from working in three research groups (robotics, neuroscience, basic biology), I’ve seen that the topic (e.g. to round out somebody’s profile) and participants (e.g re-doing experiments somebody else did so they don’t have to be included as an author, instead of just using their results directly) of a paper are often selected mainly on perceived career benefits rather than scientific merit. This is particularly true when the research is driven by junior researchers rather than established professors, as the value of papers to former is much more about if they will help get grants and a faculty position rather than their scientific merit. For example, it’s very common that a group of post-docs and PhDs will collaborate to produce a paper without a professor to ‘demonstrate’ their independence, but these collaborations often just end up describing an orphan finding or obscure method that will never be really be followed up on, and the junior researchers time could arguable have produced more scientifically meaningful results if they focused on their main project. Of course, its hard to evaluate how such practices influence academic progress in the long run, but they seem inefficient in the short-term and stem from a perverse incentive of careerism.
My impression is that questionable research practices probably vary a lot by research field, and the fields most susceptible to using poor practices are probably ones where the value of the findings won’t really be known for a long time, like basic biology research. My experience in neuroscience and biology is that much more ‘spin’, speculation, and story telling goes into presenting the biological findings than there was in robotics (where results are usually clearer steps along a path towards a goal). While a certain amount of story telling is required to present a research finding convincingly, it has become a bit of a one-up game in biology where your work really has to be presented as a critical step towards an applied outcome (like curing a disease, or inspiring a new type of material) for anybody to take it seriously, even when it’s clearly blue-sky research that hasn’t yet found an application.
As for the author, it looks like he is no longer working in Academia. From his publication record it looks like he was quite productive for a mid-career researcher, and although he may have an axe to grind (presumably he applied for many faculty positions but didn’t get any, common story) being outside the Ivory Tower can provide a lot more perspective about it’s failings than what you get from inside it.
Good point. I was commenting more on my perception of the conservation field rather than considering biases in the methodology of this study, but they keywords used were:
[insect*] AND [declin*] AND [survey]
Which does is completely biased to finding studies showing insect declines. Fig 1. also shows that most of the included studies were done in the US and Europe, with very little data coming from the tropics where most insect diversity is.
Infrastructure to support independent researchers
Epistemic Institutions, Empowering Exceptional People
The EA and Longtermist communities appear to contain a relatively large proportion of independent researchers compared to traditional academia. While working independently can provide the freedom to address impactful topics by liberating researchers from the perversive incentives, bureaucracy, and other constraints imposed on academics, the lack of institutional support can impose other difficulties that range from routine (e.g. difficulties accessing pay-walled publications) to restrictive (e.g. lack of mentorship, limited opportunities for professional development). Virtual independent scholarship institutes have recently emerged to provide institutional support (e.g. affiliation for submitting journal articles, grant management) for academic researchers working independently. We expect that facilitating additional and more productive independent EA and Longtermist research will increase the demographic diversity and expand the geographical inclusivity of these communities of researchers. Initially, we would like to determine the main needs and limitations independent researchers in these areas face and then support the creation of a virtual institute focussed on addressing those points.
This project was inspired by proposals written by Arika Virapongse and recent posts by Linch Zhang.