Watching a crow eagerly eye me for a peanut, I can’t help but wonder what it’s thinking about. Is it thinking the same thing as its flock mate, or is it having its own experience? Is it aware of me? Of itself? The conscious experience is such a fundamental part of humanity, it’s nearly impossible for most of us to envision life without it. And by extension, its hard for us to imagine that animals don’t experience consciousness too. But the fact remains that scientifically investigating consciousness, especially in non-human animals, has been slow and contentious. Among birds, this research has been all the more elusive. Which is why a study looking at subjective consciousness in carrion crows by Nieder et al. (2020)1 made an enormous splash this past fall, and resulted in a lot of misleading headlines. So why has consciousness been so difficult to study and how did this team attempt to do it?
First, let’s take a look at what we mean by consciousness. As it turns out, descriptions of consciousness get technical in a hurry, and they don’t all agree, which I suppose makes sense considering the philosophical and scientific challenge of asking, “how does an ethereal mind interact with the physical world?”2 It’s a tough question and one that can land you in an Inception-like hall of mirrors without careful consideration. So it’s critical to parse just what kind of consciousness a study is attempting to measure before making any effort to interpret their results. For the team behind this study, the focus was to determine if crows possess “subjective consciousness,” or the subjective experience of physical stimuli; in other words, the ability to have individually-specific experiences of external properties (AKA “qualia”). For example, you and I might look at a stop sign and quickly agree that it’s mostly red, but our respective experience of that redness could be quite different. A computer, on the other hand, probably does not experience qualia when detecting that an object is red, though whether or not this will soon be possible is a matter of great interest in AI circles. In addition, the study’s authors include in their definition the ability to access and report the experience of that subjective experience. For those with a background in psychology, you’ll recognize these two components as what Ned Block calls phenomenal and access consciousness.
At this point it’s helpful to step back and appreciate how much of the organismal world operates without any conceivable form of subjective consciousness. There are many a successful species that most definitely move through life by simply responding to various stimuli, without ever needing to take stock of their perceptions of those stimuli. Plants for example, react to noxious stimuli, but that doesn’t mean they have a subjective experience of pain. Even much of our own world operates without subjective experiences. Breathing for instance, happens probably 19,995 times a day without you noticing it. Highway hypnosis is another prime example. How is it that you can arrive safely at a destination you realize you don’t fully remember driving to? Because while you may not have had a subjective experience of the entirety of the drive, you were still accurately responding to the stimulus of the wheel in your hand, the pedal under your foot, and the various stimuli that presented themselves on the road.
But while it’s easy for most of us to accept that caterpillars and jellyfish probably move through life in this Simon-says kind of way, it becomes substantially more difficult to imagine that animals who look or act more like us don’t possess some from of consciousness. In fact, there are many ethologists who argue that they do, with varying levels of support.3 But the fact remains that it is difficult to demonstrate this because we cannot ask animals about their experiences and perceptions. We might, however, be able to leapfrog the inconvenience of working with nonverbal subjects by going directly to control center itself: the brain.
Among primates, including humans, the brain neurons that are responsible for representing what an individual perceives (i.e subjective experiences) are in the neocortex, which is part of the mammalian pallium.4,5 Until relatively recently, replicating such studies in birds was ignored, because bird brains are unique from mammalian brains in some key ways. As a result of these differences, we used to believe that birds had no equivalent to the mammalian neocortex, and were therefore incapable of flexible, complex thought. Now we understand that the circuits of the avian pallium are functionally organized in a similar way to the mammalian pallium, and furthermore that the avian pallium contains a staggering density of neurons.6,7 This is especially true among the corvids and parrots, which like non-human primates, can have as many as 1-2 billion neurons in their pallia.8 Given this organizational similarity, we’ve been able to show that birds do in fact have an analog to the mammalian prefrontal cortex, the part of our brain the allows us to process thoughts, feelings, and decisions. The question now is whether this avian analog, called the nidopallium caudolateral, also houses neurons that allow for subjective consciousness.
If any part of that was lost on you, we can summarize it as follows: the reason you and I can experience the exact same stimulus (the color green, the prick of a needle, the sound of C-sharp) differently, is because we experience subjective consciousness. Philosophical/religious discussions aside, the experience of consciousness is regulated by the brain, and we believe we’ve identified the brain areas and neurons that control this in humans and some primates. The goal of this study was to look to the analogous area of the bird brain where these neurons are housed in the primate brain, and see if they could demonstrate a neuronal basis for subjective consciousness in birds.
In theory that sounds simple enough (lol), but how exactly did they use neuronal activity to identify the experience of subjective consciousness? The fundamental “trick” of this study is that they exploited a stimulus that can give rise to two different percepts. Think of the classic optical illusion of the rabbit-duck. Whereas what you see first is a rabbit looking to the right, I see a duck looking to the left, and this difference is evidenced by the respective activity in our brains. This study used that same concept only instead of optical illusions, they used light.
By implanting electrodes in two carrion crows’ brains and training them to report if they saw a light, the researchers could investigate their subjective experiences by reducing the intensity of the light to near perceptual threshold levels and asking the crows to indicate whether they had seen it. Importantly this training wasn’t as simple as, “peck if you saw a light.” Instead, after the light was presented, there was a pause, after which the bird was shown either the color red or blue. Depending on whether the bird thought it saw the light or not, which color it was then shown informed how it was supposed to indicate its perception to the researchers. If the light had been detected, seeing the blue color meant it would only get a treat if it stayed still, whereas red meant it needed to move. If that light hadn’t been detected, then seeing blue meant it needed to move, while red indicated that the bird should stay still. Without this step of making the birds wait to find out what motor response indicated their answer (and earn them a treat) the study would only have revealed the neurons associated with preparing the correct motor response. Instead, they were able to look at the neuronal activity related to the immediate impact of the stimulus, and then to the activity related to processing this information into a perception.
What they found is that like primates, crows exhibit a two-stage process, where neuronal activity during Stage I mostly reflects the intensity of the physical stimulus, followed by a second spike in activity that reflected their perception. The patterns of activity in Stage II were so consistent, that the researchers could predict whether the crows would say they saw the light or not by looking at this activity alone. Most importantly, while the responses of the two birds were the same if the light intensity was bright and unambiguous, when shown faint lights, the two birds responded differently. Meaning that despite being shown the exact same stimulus, the two birds had different subjective experiences of whether they had seen it or not. There were also instances of false positives, where the birds indicated that they had seen a light that wasn’t really there. In these cases their brains behaved during Stage II just as they did when they had actually seen a bright light. This is important because it further demonstrates that the brain activity the researchers were measuring correlated with the crows’ subjective experience, rather than as a result of the intensity of the stimulus itself.
What this shows us is that carrion crows have the neurological substrates that support subjective consciousness, and it indeed appears that they have individual experiences of stimuli. It does not show us, despite many articles to the contrary, that they are “self-aware” or engage in metacognition (the ability to “ponder the contents of their own minds“). Still, these findings makes crows pretty unique among animals, putting them in a category shared only by primates. Furthermore, it underlines that despite the differences between mammalian and avian brains, the two are are remarkably functionally analogous, at least with respect to some species. In fact some have gone as far as to say that this and other studies indicate that the continued assertion that birds do not have a cerebral cortex is outdated and wrong.8 Moving to the 30,000ft view, the findings of this study invite numerous questions about what such shared abilities say about the evolution of consciousness across species. Did it evolve independently multiple times or has is been present since before the evolutionary split between birds and mammals some 320 million years ago? Either way, if I was a betting person, I would wager that the list of animals in possession of subjective consciousness will only continue to grow as we find new ways of exploring these once out of reach questions.
~Many thanks to Dr. Andreas Nieder for helping me parse the methods and findings of this fascinating and complex paper.
- Nieder A, Wagener L, and Rinnert P. 2020. A neural correlate of sensory consciousness in a corvid bird. Science 369: 1626-1629
- Glatterfelder JB. 2019. Subjective consciousness: What am I? In: Information—Consciousness—Reality. The Frontiers Collection. Springer, Cham. https://doi.org/10.1007/978-3-030-03633-1_11
- Boyl M, Seth AK, Wilke M, Ingmundson P, Baars B, Laureys S, Edelman DB, Tsuchiya N. 2013. Consciousness in humans and non-human animals: recent advances and future directions. Frontiers in Psychology 4: https://doi.org/10.3389/fpsyg.2013.00625
- Dehaene S. and Changeux JP. 2011. Experimental and theoretical approaches to conscious processing. Neuron 70: 200-227
- de Lafuente V. and Romo R. 2005. Neuronal correlates of subjective sensory experience. Nature Neuroscience 8: 1698-1703
- Shanahan M, Bingman VP, Shimizu T, Wild M, and Güntürkün O. 2013. Large-scale network organization in the avian forebrain: a connectivity matrix and theoretical analysis. Font Comput Neurosci 7: doi: 10.3389/fncom.2013.00089
- Olkowicz S, Kocurek M, Lučan RK, Porteš M, Fitch WT, Herculano-Houzel S. and Němec P. 2016. Birds have primate-like numbers of neurons in the forebrain. PNAS 113: 7255-7260
- Herculano-Houzel S. 2020. Birds do have a brain cortex-and think. Science 369: 1567-1568