How can we tell if an animal is depressed?

CHRIS HATZIS
Eavesdrop on Experts, a podcast about stories of inspiration and insights. It’s where expert types obsess, confess and profess. I’m Chris Hatzis, let’s eavesdrop on experts changing the world - one lecture, one experiment, one interview at a time.

How can you tell if an animal is depressed? Well, if your pet rat has long periods of staring into space, doing nothing and not eating much, that could be one indication. But what if you're looking after an octopus - what do we need to understand about an animal's brain functions before we start assuming that their mental state isn't quite up to scratch? It may seem like a bizarre question to philosophise about, but the answer fundamentally affects how animal welfare is legislated.

MIKE MENDL

I'm Mike Mendl. I'm Professor of Animal Behaviour and Welfare at Bristol University in the UK.

CHRIS HATZIS
Professor Mike Mendl is set on ensuring the welfare of all creatures big and small. He was in Melbourne recently to present a public seminar at the Animal Welfare Science Centre titled “Assessing Animal Affect: How can we tell what a domestic animal is thinking or feeling, drawing on animal behaviour, human psychology and cognitive neuroscience.” Here at the University of Melbourne he spent some time chatting about his work with our reporter Dr Andi Horvath.


ANDI HORVATH
Now if I met you at a vegan barbecue, how would you describe what you do?

MIKE MENDL
I'm a researcher in animal welfare, so I work on the welfare of animals. Trying to develop ways of measuring their welfare, in particular scientifically and with a general aim to try and improve animal welfare. Also improve productivity or improve the quality or conditions for animals and this could be laboratory and farm animals, and companion animals too.

ANDI HORVATH
How have humans changed their thinking about animals over the last centuries?

MIKE MENDL
That's an interesting question. Obviously, there's been a lot of different views that people have had about animals. The animal rights and animal welfare organisations came into being a couple of hundred years ago in the UK. These have, I guess, focused on animals as being sentient organisms, who are capable of suffering, and maybe changed somewhat how we view them and how we should view how we treat them and keep them. So, there has been a shift in that way, which is not just that reason, but there's been, certainly in the last couple of hundred years, a shift towards that. I think, as a result, there's now currently, in countries like the UK in particular, a very big concern about animal welfare and variety of different contexts in which animals are kept.

ANDI HORVATH
What changes have you seen, since you first entered the field, that's already shifted animal welfare?



MIKE MENDL
My field is animal welfare science, so we try and develop ways of measuring animal welfare. Also then using those methods to try and see what animals are experiencing in terms of welfare in different environments and how we might change the practices to improve their welfare and productivity at the same time. I think there's been a shift from thinking of animal welfare measures as being partly, or mainly, to do with the biological functioning in terms of physiology, reproduction, injury and so on to a concentration more on the mental states of animals. Acceptance that animal welfare, for many people, is a concern about whether animals are able to suffer or feel pain, thirst, hunger and therefore we need to know more about their mental state.

ANDI HORVATH
Invite us to follow you round for a couple of weeks, like an intern. Explain a case study for us that you're working on, that examines particular animals. What animal do you study? What are you measuring?


MIKE MENDL
We're doing quite a lot of work on laboratory animal welfare currently in the UK. Some of our projects involve looking at quite simple behavioural measures, for example, measures of activity or inactivity. For example, time spent with the eyes open but doing nothing, which one of my colleagues, Doctor Carole Fureix, who's at Plymouth University, has proposed to be a useful indicator of depression-like states in mice and other animals. So, part of our work would be to look at whether we can correlate how much time animals spend lying around with their eyes open, which seems to be equivalent to apathy, depression-like state, and correlate that with other indicators of welfare. To see whether it might be a good indicator of this particular state.

ANDI HORVATH
Laboratory animals have been used for quite a few decades. Have there been changes there?


MIKE MENDL
There's been lots of changes in legislation. Again, speaking from a UK perspective, in 1986 there was a major Act, the Animals (Scientific Procedures) Act, which came in to try and safeguard animal use in laboratories and made getting licences to do work on animals a much more rigorous process. That's developed over the last few years again, with some changes in the last decade, which have increased the number of different sorts of species, which are covered by this Act and so on.


ANDI HORVATH
There's quite a range of laboratory animals. Do they still use guinea pigs?



MIKE MENDL
Oh yes, guinea pigs are still one of the animal species. The most commonly used species is still the mouse, laboratory mouse. Fish, zebrafish in particular, are becoming much more commonly used now, and rats. So, mice, rats and zebrafish make up at least in the UK again, the majority of animals used. But there are a lot of other species used in much smaller numbers.

ANDI HORVATH
It's quite a controversial area, or has been, rather. Tell us about your experience of the controversy involved in animal use in laboratories.

MIKE MENDL
Well clearly there has been a lot of heated debate about the use of animals in experiments. Why they're being used, what they're being used for, what their experiences are in those environments. Again, in the UK there has been a lot of - in the past, not so much recently, but in the past, quite a lot of activity with direct action, including bombs and so on, which obviously have made laboratory animal researchers' lives a bit more uncomfortable.

Currently, there is a push towards being quite open about what goes on in laboratories and I think that's a good thing, if people can do that safely. From our point of view, we obviously look at the way in which the animals are kept and whether that can be improved in some ways through changing caging. Many of the companies who build cages are also looking into much more innovative ways of caging, giving animals more space, allowing them to access more than one cage and so on. So, there are lots of developments going on in that area.


ANDI HORVATH
Is research restricted by our assumptions about what an animal may be thinking and feeling in terms of its ability to thrive or its vitality?



MIKE MENDL
Yes, I think that underpinning all concerns about animal suffering and welfare is some assumption about the ability of the animals to experience suffering, to experience pain, to be frustrated. To be simple things like hungry, I guess, and if not fed well. Those things are much less of a concern usually because the animals are usually fed very well. But all these mental states are things that animals are assumed to have, those animals which we have legislation to protect. I guess the cut-off lines perhaps somewhat reflect the assumptions that some species, we're not so sure about their abilities to suffer or to have these experiences and therefore we don't legislate to protect them. There are obviously controversies about where that cut-off line should be made. Currently it's cephalopods in the UK, they're the main invertebrates, octopuses, squids and so on, they're the main invertebrates who are protected under the laboratory animal legislation. Whereas insects and many other invertebrates are not.

ANDI HORVATH
So, some animals have a psychology that is, “I've got to eat, I don't want to get eaten and I've got to get a mate.” But clearly with things like octopi, there's something more involved in their psychology.



MIKE MENDL

Well we think there is, so, obviously, as you say, all animals have to do these, vital survival activities are finding things to eat, avoiding being eaten. But animals vary in the social organisation they have, and they vary in the sorts of tasks that they require to do to get their food, some of which require much more complex behaviours than other sorts of food acquisition. So, octopuses may have to catch organisms, which they can then open and extract food from and they maybe need to learn particular associations, which help them to do this.

They also have some very complex camouflage abilities, which are - again, the cognitive processing involved is not entirely clear. But people are looking into what sort of processes are required for them to assume the shape and appearance of objects in amazing ways. They do some very clever things, to our eyes. Whether that then reflects that they are particularly intelligent or consciously aware in the way that we are, if we were to do these things, is another question. That's something that's very difficult to answer directly because we can't get into the heads of these other animals directly.

ANDI HORVATH
Just as an aside, is there truth to the common knowledge that octopi, and I think it's crows, have larger brain size compared to their body?



MIKE MENDL
Well yes, so the brain size relative to the body is one thing, and certainly certain species have a larger encephalization index. The cortical parts of the brain are bigger than one would expect and the brain in total is bigger than one would expect for the body size. So, you can draw a line looking at body size against brain size and there are some outliers, where the brain size is much higher than you would predict, for body size. But another thing that's coming to the fore now is the understanding of neuronal density.

So, bird brains have much higher density of neurons and they cram many more neurons into a small space than an equivalent mammal. That makes a lot of sense for an animal, which has to do complex things, but also has to fly so it wants to have light apparatus on board. One way of doing this seems to be to have a much more densely packed neuronal structure and architecture.


ANDI HORVATH
Oh gosh, we've got to stop using the word…

MIKE MENDL
Birdbrain.

ANDI HORVATH
…you're a birdbrain…



MIKE MENDL
Exactly [laughs], or use it in the other way.

ANDI HORVATH
…to describe someone as not being quite clever. I'm keen to understand the role of evolution and genetics in this. So, let's start with evolution. How has this informed animal welfare?

MIKE MENDL
That's a good question. Because a lot of our animals who we are working with directly, in terms of improving welfare, are, for example, domesticated animals, farm animals and so on. So, evolution there is a complex thing because we'll have the process of domestication. That has obviously affected animals in some ways, in the sense that they have become much less fearful of people than they would have been. So, they are certainly able to deal with things that a wild animal would find extremely challenging. We have to take that on board when we look at how we might improve welfare and acknowledge that they can certainly withstand being closer to people. For example, having contact with people on a regular basis, compared to a wild ancestor. So, there are things like that, which obviously affect how we perceive those animals.

ANDI HORVATH
That was actually going to be my next question. Can the genetics of domesticated animals, compared to the wild, give us clues about the behaviour? You've just said, yes, it does, they're used to us.

MIKE MENDL
Yep.

ANDI HORVATH
So, how do we then approach animal welfare? I want to ask you, is it a cost/benefit analysis for humans, for society, for animals? Is this comparative weighted trade-off actually a way of thinking that is even appropriate?

MIKE MENDL

I think we can approach - in terms of what the animals require, for example, so our baseline might be something to think about what the animals might want in their environments and whether we're able to provide that. Then obviously genetics, as I've said, and evolution has changed what they require. But there is a lot of evidence too that domesticated species, if, for example, released into the wild, behave in quite similar ways to their ancestral species. Feral pigs, for example, will form groups, which are very similar to the wild boar ancestor, and so on. So, you will see quite a lot of similarities there.

So, there are still certainly thoughts that they could require and could, certainly, benefit from particular sorts of environments, which their wild ancestors would be in. Albeit that they are much more used to human presence and less affected by it. So, there are those issues, which we need to think about and means that we could assume that these animals have quite similar requirements to ancestral species.

But what we actually do is, we try and measure what they want. So, part of animal welfare science is to measure what animals want, by seeing how hard they work to access particular resources. A particular sort of - it could be something as simple as a cage floor type. Do they prefer this type to that type? We can ask them questions like that and see what is the most preferred environment. With the assumption that if they get the preferred things, they are going to be in a more positive emotional state.

They're going to be in a state that they want and that will be better for their welfare. One can then design from first principles environments. Of course, they may not be suitable sometimes, they may be difficult to commercially employ. So, one has to then see how one can they integrate these things into a commercially affordable system. That's one way, certainly, of trying to improve animal welfare.


ANDI HORVATH
You've made me think of chickens. I know that there's been research done in the types of floors they prefer.

MIKE MENDL
Yeah, yeah.

ANDI HORVATH
Because they like to scratch…

MIKE MENDL
Yes.

ANDI HORVATH
…and have dust baths and things like that. So, these sorts of changes have been implemented in some of the egg-laying industry.



MIKE MENDL
Yeah.

ANDI HORVATH
I'm fascinated by this discussion of being re-released into the wild and that they've still got their ancestral genetics and behavioural capabilities to survive. I'm not sure if I was let out into the wild, I could survive…



MIKE MENDL
No.


ANDI HORVATH
…as humans.



MIKE MENDL

I think you're right. That animals released into the wild, often inadvertently or escaping from farms or whatever, they are probably less able to avoid predation and so on, because of their background experience. I think after generations, if they survive for generations, in the wild, they will develop these, or re-develop, these abilities. Certainly, the animals - it will be some sort of selection procedure going on then, those animals who escaped, who were good at avoiding predation, are likely to give rise to the next generation. Initially, when a domestic species gets into the wild it will be very vulnerable.

This is the same also with wild animals who have been raised in captivity. I'm involved in some work, which is looking at the welfare of rehabilitated primates in, for example, South America. It's clear then that, unfortunately, some of these animals are used in the illegal pet industry and they come in as wild animals, who have been kept as pets. They need to be retrained, really, to adapt to the challenges of being in the wild again. So, you can see in an individual's lifetime, their experience can change how well they can cope with the wild environment. But they have the ability there, which, if retrained, can be re-introduced.

Likewise, even for domesticated species in their evolutionary lifetime, they've probably been less exposed to the pressure of predation, but there seems to be something there which will help them to survive if they are released.

ANDI HORVATH
There's something said for then learning SAS and scouting skills [laughs]…



MIKE MENDL

[Laughs] Yes.


ANDI HORVATH

…to survive in the wild. Whilst you've been talking about animals, I can't help but think about individual variation in animals. As a dog owner, I've had many dogs and they've all had different personalities. They've all had different, what as a human, I subscribe to as a cheeky behaviour, different types of cheekiness. How do you resolve this issue of individual aspects of an animal's psychology?

MIKE MENDL

It's a very good question. Because, obviously, animal welfare changes are often made for the group or for the population of animals and there will be variation in how the animals deal with the environment. Just the social environment, obviously some animals are likely to be higher up the pecking order than others when in groups which have some sort of social organisation like that. Therefore, some animals will be exposed to aggression and so on from other animals and do less well in the same environment. So, one can then think about ways of trying to make provision for those animals to escape attack or so on. But it is the case, obviously, that the environments one builds may not suit all animals to the same way because there is great individual variation in behaviour.

To try and overcome that, as I said, the thing to do is to try and build in features, which some animals, who may not otherwise do so well, can take advantage of. Areas to escape from others and so on and make sure that resources are all evenly spread so there's no bunching of a resource, which means that only the top rankers get access or whatever. So, there are things one can do to try and even out the inevitable disparity between individuals in the same group.

ANDI HORVATH
Over the centuries, as you've said, we have redefined sentience, with reference to suffering, but also capabilities and intelligence. Have we redefined intelligence enough for animals?



MIKE MENDL
I think intelligence is a really difficult question for animals, because each animal has adapted to a different niche. So, within its own niche, it can appear to be very intelligent. It can do things, which are what we might think of as not particularly clever from our point of view, but actually are very well adapted to that environment. So, defining intelligence in a very general way is not easy. Some people fall back on a definition, which is to do with flexibility and the ability to actually tackle lots of different issues.

Being a generalist in a sense, have quite complex cognitive abilities that allows you to tackle a range of different issues. That may be something we can think of as intelligence. But in other ways, being able to detect sonar, being able to remember where you've hidden 10,000 nuts or whatever, these are things that animals can do and they're different sorts of intelligence - very good memory, very good sensory abilities.

So, intelligence itself is quite a difficult concept, particularly to compare across species it's difficult. But certainly, some species are much more adaptable, flexible, can do a variety of different things. Have a complex organisation socially, so it can do very complex social things, which people often of as being very intelligent. So, they can deceive each other apparently or predict what each other's going to do, and these things are often seen as 'intelligent'.


ANDI HORVATH
That's right. Problem solving is actually based on the context. Just like there are various CEOs that are appropriate for the right stage of an organisation, whether it's a start-up or whether it's a re-invention or whether it's - so problem solving in humans is a huge spectrum as well. So, it's no surprise that it's a spectrum in there. I want to go back to the tricky topic of animal welfare. Some of the assessments for animal welfare, obviously are scientific. There are measurements, there are ways of understanding. But it becomes a morality question as well. How do you assess the morality of what humans are doing with animals in society?



MIKE MENDL
Our work is primarily at the scientific end, so we're trying to make scientific measures of welfare, which we can then use as a metric for the effect of what we're doing on the animal. But you're right, that obviously the ethical judgements then of what we do, and the costs and benefits are something which, I guess, broader society has to also be involved in.

Scientists, in some ways, animal welfare scientists at least, are providing data on the effects on the animals. So, when it comes down to the ethical issues of using animals for experimentation, for example, for eating, what are the costs and benefits, then it is a societal question. But certainly, philosophers can have input and scientists can have input.

Generally, for example, in laboratory animal welfare, there's an idea of trying to trade off - my PhD supervisor, Sir Patrick Bateson, who recently sadly died, he was very well known for his so-called Bateson's Cube, which traded off different types of consideration - the quality of the science, the suffering of the animal and the medical benefits.

For example, you could trade that off and make a sort of utilitarian judgement, to a certain extent, about when one could justify what sort of experiment. That's just one approach. Obviously animal rights people have a different view. They think absolutely we should not be using animals in these contexts. So, there are different philosophical standpoints and it's a very complicated issue, with lots of different people's views.


ANDI HORVATH
At least we have some data to fall back on, that asks sensible questions about animals.



MIKE MENDL
Yes, we hope that animal welfare science measures - I mean they're not perfect, of course, but they are at least providing some objective data to the debate. Which is really a major role that animal welfare science can play, together with providing ways in which one can implement things and improve welfare through change in environments or whatever for the animal.


ANDI HORVATH
What are some of the common misconceptions you've encountered, be it public or even scientists, about animal welfare?


MIKE MENDL
It's an interesting question. Some people are just surprised that one studies animal mental states or even are particularly surprised that certain species would be considered to be sentient. There are still arguments about fish, for example - oh, they don't feel pain, do they? So, there's that sort of attitude amongst some people. Other people are at the opposite end of the spectrum and saying, what about insects? Should we be concerned about insect welfare? There's quite a lot of discussion ongoing at the moment about what insects might experience, very interestingly.

I guess, a lot of people are concerned about animal welfare and they are maybe surprised about how animals are kept, in some circumstances. Not aware of that and what goes on and therefore keen to see, to understand how one can assess the effects of those things on the animals and how one might improve welfare.


ANDI HORVATH
I met an animal activist once, who demanded to see how the animals were kept. Years later, she ended up working in the area, which was a very interesting transformation. She really wanted to understand it and get first-hand knowledge. Not just push a barrow but be open to what's really going on and what could be better.


MIKE MENDL
Yeah, I think there are quite a lot of people who are just interested generally in how the products that they use are raised really and kept. They welcome an insight into that.

ANDI HORVATH
What surprises have you encountered in your research field?



MIKE MENDL
That's a good question. A big area of our research has been trying to develop new ways of assessing animal welfare, as I've said. One thing we've been interested in, is the links between an animal's effective state or emotional state and the way it makes decisions and its cognitive and decision-making behaviour. I guess our stance is that animal welfare is about what animals experience and feel, so it's about emotional states and effective states in animals. Of course, we can't directly access those because they're private and of course some people would argue that maybe these states don't exist in certain species. That's another philosophical argument.

But we've been interested in seeing whether, like humans, animals show links between how they're feeling and how they make decisions. So, one of the areas we've worked in is the area to do with decision-making in ambiguous situations. There's lots of data from human studies that people who are unhappy, depressed and so on, tend to be much more pessimistic about the future and they make more careful, cautious decisions about ambiguity, compared to people who are in a more positive, happier state. We've been looking as to whether there are similar relationships between decision-making and effective state and welfare in animals. So that we can use measures of decision-making as a proxy measure of emotional state, and hence welfare.

ANDI HORVATH
I know when I’m feeling flat, I can't make decisions. Like, there's a banana and an orange and I go, what do I need? What do I really need? What do I really feel like? I don't know. Oh, I should eat that, no, I haven't had enough of that. It's quite a turbulent [laughs] thing for a simple decision when you're flat and tired.

MIKE MENDL
Yeah.

ANDI HORVATH
Are these the - are they food decisions? Or what do you put in front of the laboratory animals?



MIKE MENDL
The way we look at this question, is we train them on a simple task, which is to discriminate a cue, for example, a sound which predicts something nice, like food, a bit of food arriving, and another cue, a sound of a different frequency, which predicts something less nice, like a nasty tasting food arriving or no food arriving or a puff of air or something like that. The animals learn to respond to these cues in appropriate ways. So, for example, they will approach the feeder to get the food when one cue, the positive cue, sounds. When the other cue sounds, they will avoid the feeder and just keep away. So, they learn this discrimination and then we can just simply, occasionally, then ask them how do you judge an ambiguous intermediate cue? So, a cue of a different tone, which is different to the two training cues, is ambiguous to them and we say, do you approach it? Are you, if you like, optimistic? Or do you stay back? Are you pessimistic? Then we try and see whether that relates to the animal's welfare state. So, if animals who we think are in a more negative environment, for example, make more pessimistic decisions, that's the way we've looked at it.

ANDI HORVATH
Do animals have cognitive biases? They obviously clearly have preferences and automatic responses. How do you make sense of that?

MIKE MENDL

Yeah, so the work we've been doing is on this decision-making under ambiguity, which would be what we would call a cognitive bias. In the sense that the animal is appearing to make a decision, which is biased by something else. In this case it's an emotional or effective state, and that's a method we've been using to try and assess animal welfare. Basically, if the animal is showing a pessimistic, quotation marks, cognitive bias, then we could interpret this as the animal being in a more negative state. If it shows a more optimistic bias, we interpret it as being in a more positive state. We're doing a meta-analysis now on the hundred studies which have been done using this approach now, to see whether that's a robust finding. It seems to be reasonably robust, but there are exceptions, as ever one would expect in science and we need to understand what those exceptions are about and what they mean.

ANDI HORVATH
What inspired you to enter this field? Was there a particular case study that moved you? Was it a book? A professor? A pet?



MIKE MENDL

Moving into animal welfare as a whole field, I think that came from doing animal behaviour as a PhD in Animal Behaviour Studies. Then thinking more about, how I apply this really and becoming more aware of the growing field of animal welfare at that time. My professor, Professor Donald Broom, in Cambridge, who ran an animal welfare group - I was doing a postdoc in Holland at the time and became more aware of animal welfare as an area where one could apply animal behaviour science. I contacted Don and had an interview with him and luckily got a job and that's when I moved into the more applied area of animal behaviour, which is looking at animal welfare.

In terms of the work we've been doing on cognition and welfare decision-making and welfare, that's - through particularly my partner, Doctor Liz Paul, who's a psychologist, who was very aware of the human literature on the links between people's mental states and how they make decisions. So, we, together, explored the possibility of developing this approach in animals.


ANDI HORVATH

Oh, you met a psychologist and that's what…ah.



MIKE MENDL
Yeah, in terms of that approach, has come through the understanding of human psychology. In fact, one can argue that if one believes animal welfare is about feelings that animals have, then the best model of that is actually humans. Because humans, we can talk to each other about our feelings. We have a more direct route to understanding somebody's feelings and therefore we can look at the links between how they're feeling and their physiology, for example, or their behaviour or their decision-making, and attempt to translate that to animals. Of course, being careful not to do it in an uncritical way.

ANDI HORVATH
I'm often amazed, but not surprised, that research is often driven by meeting someone in an allied field, or an interdisciplinary area or a cross-disciplinary area, and that's what you have done.

MIKE MENDL

Yeah.


ANDI HORVATH
I'd like to ask you, next time we see a mouse, or we hear about laboratory animals, what would you like us to think?


MIKE MENDL

I think I'd like you to think about how complex that organism is. Even if it's a drosophila, so a fruit fly, they can do very complicated things. I've been working with researchers who work with fruit flies and they do very complex decision-making tasks, learn things. Likewise, the mouse has a potentially very complex life, particularly its olfactory life, which we're very unaware of. The smells that many mammals use in their day to day existence and the way in which they use that information is very complex.

So, it's a complicated organism in front of us, doing something - living a complex life. We have to think about then, how we take that into a more confined laboratory environment. How that affects it, how that affects the findings we get from it. So, there's obviously concerns that animal welfare improvements in laboratory animals also improves science because the animals are in a better state.

ANDI HORVATH
Professor Mike Mendl, thank you, on behalf of us, and the animals.



MIKE MENDL
Thank you. It's been a pleasure, thank you.

CHRIS HATZIS
Thank you to Mike Mendl, Professor of Animal Behaviour and Welfare at Bristol University in the UK. And thanks to our reporter Dr Andi Horvath.

Eavesdrop on Experts - stories of inspiration and insights - was made possible by the University of Melbourne. This episode was recorded on May 15, 2019. You’ll find a full transcript on the Pursuit website. Audio engineering by me, Chris Hatzis. Co-production - Silvi Vann-Wall and Dr Andi Horvath. Eavesdrop on Experts is licensed under Creative Commons, Copyright 2019, The University of Melbourne. If you enjoyed this episode, drop us a review on iTunes and check out the rest of the Eavesdrop episodes in our archive. I’m Chris Hatzis, producer and editor. Join us again next time for another Eavesdrop on Experts.