Tuesday, December 8, 2009

Non-human Primate Models of Psychiatric Treatment Effects

Before starting the main body of my post, here's a little introduction:

I've been doing quite a bit of reading lately about the history of psychiatry (in particular, an excellent book by Lisa Appignanesi; I'll write a post about it when I've finished, which could be in a while, since the book is 5 cm thick!). Also I've been reading about cultural psychology (another very interesting field), after finding a free set of university lecture notes published online. I'd like to write another post about this subject as well, when I get around to it.

What does this have to do with "non-human primate models of psychiatric treatment effects?" Well, I'm becoming more strongly aware of the powerful effects of culture upon the manifestations of psychological (and, possibly, physical) health and distress. The book I'm reading deals with cultural change through history; these changes have influenced the presentation, management, and course of many psychiatric phenomena. Even terms like "psychiatric phenomena" or "symptoms," etc. are culturally influenced jargon. The cultural psychology subject also deals, of course, with cultural differences, but in this case mainly with the way different groups of people in the present era around the world experience or perceive emotions, psychological distress, social interactions, or cognitive processes. I suspect that cultural differences may exist between families as well, within the same geographical area.

These factors complicate the study of psychiatric therapies, perhaps in many ways that could be subtle but powerful.

I've been interested in finding more evidence about the effect of physical and psychological treatments for psychiatric symptoms in non-human primates. In this case, cultural or personal history biases could be much more carefully controlled.

There are a lot of studies done in rodents, of behavioural therapies and of medication, including a very questionable rodent "model" of antidepressant effectiveness. I think that possible conclusions are much more limited, about human therapies based on research done in mice, etc.

Monkeys or apes are much closer to humans, in terms of genetic similarity and brain structure. They may exhibit behavioural problems that are much more closely analogous to psychiatric symptoms in humans. So, I have been looking for good research about medication and "psychotherapy" effects in primates. Here's a start:


This 2009 article describes self-injurious behaviour in rhesus macaques. These animals may bite themselves severely; this is thought to be due to an underlying vulnerability combined with social deprivation in infancy or being isolated in captivity. About one-third of macaques experiencing solitary captivity exhibit self-directed stereotypic behaviour. The behaviour is exacerbated by separation from the social group, by disruption of daily routines, or by exposure to a fear-provoking stimulus (for animals, this could be an unfamiliar person trying to interact with them closely). It is interesting to consider that analogous behaviours in humans are probably related to similar vulnerabilities, deprivations, or triggers.

The experiment described in the article is about treating these self-injuring monkeys. Each group started off with 4 weeks of baseline observation, followed by 4 weeks of placebo, before randomization to fluoxetine, venlafaxine, or placebo for the final 4 weeks.

The individuals in the fluoxetine groups, at higher doses in particular, had substantial reductions in self-injurious behaviour (at least 50-75% less self-injury than the placebo group). The venlafaxine group did not improve as much.

There were no changes in "general behaviour" aside from a reduction in "aggressive displays." In particular, there were no signs of sedation or reduced engagement, etc.

I don't mean to make too much of results of this type, but I do think that this is strong evidence that the effect of an SSRI is not simply of an elaborate active placebo, influenced by cultural expectation. Also, just because a symptom is reduced doesn't necessarily mean a problem is solved...however, reducing a problematic behaviour such as self-injury may be a necessary prerequisite to resolving other types of psychological problems.

This type of study would be strengthened if it was extended for a year or more, and if it was to include data about other "quality of life" indicators, such as social integration, longevity, physical health, etc.

Here's another study, showing that tryptophan administration over a 4-week period substantially decreased self-injurious behaviour (again, by 50-75%) in small monkeys. There was also a decrease in previously high levels of cortisol. The dose of tryptophan was over 100 mg/kg per day, which would be a bit inconvenient to administer to humans:

Here's another study of self-injury in macaques. In this case, housing the animals outdoors led to significant reductions in self-injury. I think the message here could be that a healthy environment which optimizes freedom of movement, space, and natural sensory cues (e.g. of light, sound, and temperature), leads to diminished stress and and diminished symptoms of psychological distress. We could confidently generalize this statement to humans, I think.

Here is a relevant review on the subject of self-injury in human vs. non-human primates:

Here's an amusing (and, unfortunately, not very strong) study showing that hearing music leads to increased affiliative behaviour and decreased aggressive behaviour in chimpanzees. There were different degrees of responsiveness to different types of music:

I'll try to add to this post later. In the "psychotherapy" realm, some of the first important animal studies in primates were done by Harlow. I'm interested to find some more recent stuff in the research literature. I guess there won't be much on cognitive therapy in monkeys, since there is a bit of a problem encouraging non-human primates to keep written diaries with thought records...similarly, psychoanalytic studies are probably in short supply (!) Yet, in all seriousness, I suspect that the key elements for successful therapy in non-human primates involve positive, gentle, consistent relationships; and gentle, non-punitive behavioural education & modeling.


Anonymous said...

The obvious concern with primate studies is in regards to ethical practices, such as may arise when inducing psychological trauma. The study you cited obtained their subjects from the wild, but then one can argue if treating the SI behaviour outweighs removing the animal from their social group.

With respect to reducing cultural bias, we need to consider the bias of the observer/experimenter. As well, to use primates as a test subject similar to humans, the closer we get to a human-type animal, the more we are going to encounter similar limitations. Specifically with reference to culture, primates (especially the higher order ones) do have cultures of their own. I know there has been a lot of research into primate behaviour and I would be interested to see if we can control for cultural differences between geographically different groups.

Overall, I think we need to seriously consider if there is anything significantly new we can learn from primate studies. Maybe all we need to do is come up with novel ways of analysing and interpreting data, as well as new experimental designs.

GK said...

I think it is abhorrent to deliberately induce psychological trauma. My posting is in no means intended to affirm such a practice. I agree that we can come up with novel research techniques which do not involve causing any new suffering.

However, part of my point here is that creatures other than humans also experience psychological distress, either as a product of innate vulnerability, or environmental adversity; this distress may manifest itself in a variety of ways.

I consider it valid, useful, and highly ethical to consider how to help an animal in distress.

The elements of culture I am particularly concerned about, with regard to psychological therapies, have to do with cognitive constructs and abstract ideas about emotion and about therapy. While there is much to understand about the cultures of different primate species, I think most would agree that intellectual abstraction about psychological phenomena and about current trends of psychological therapies are unlikely to be factors characterizing non-human primate life.

In particular, the impact of a placebo would have to be very substantially less in a non-human primate, due to the reduced ability such creatures have for appreciating therapeutic strategies in an abstract or intellectualized manner (a manner which may be highly influenced by the intellectual culture and abstract beliefs in human societies). A placebo effect may certainly manifest itself in animals if the human administering the therapy is unblinded, since that human may be biased to treat the active subjects differently than the controls, at least on an unconscious level.

I think that part of a therapeutic project ought to include ways to help ALL creatures, including our non-human relatives. Not a lot of effort is spent trying to help animals psychologically. Therefore, I do certainly think there is much we can learn that is significantly new from primate studies.