Experiments at The State University of New York Health Science Center

The relevance and importance of maternal deprivation monkey experiments continue to be scientifically debated because of conceptual and methodological flaws in the experimental design. Psychologist and monkey researcher Leonard Rosenblum has continued this research paradigm to investigate human panic disorder through an induced condition in monkeys, believed to sufficiently replicate panic disorder to quality as an animal model. However, since panic disorder cannot be diagnosed in monkeys by established clinical psychiatric criteria, and since specifications for a valid animal model established by monkey researchers themselves are not met, the monkey condition does not model the human one. Critical psychological factors correlated with panic disorder are not convincingly demonstrated in monkeys, and laboratory-induced stress confounds the results. Therefore, as is typical of maternal deprivation experiments, Rosenblum's non-explanatory monkey data merely dramatize already known human findings by unnecessarily transforming ideas from one conceptual system (human psychology) to another (animal behavior). For reasons of economic feasibility, scientific validity, and medical accuracy, Rosenblum's animal studies in particular, and maternal deprivation monkey research in general, should be defunded and phased out.

For over 30 years, researchers have forced monkeys to undergo maternal separation or deprivation in order to "model" such human conditions as depression, alcoholism, aggression, and maternal-infant bonding.^1-15 Three major critiques by Michael Giannelli,¹⁶ Martin Stephens,¹⁷ and Brandon Reines^18-19 have challenged the scientific underpinnings of this work. An exhaustive review of the literature has not revealed a single attempt by the monkey researchers to address criticisms raised by these scholars. Expanding on the work of Giannelli, Stephens, and Reines, this report focuses on the research of Leonard Rosenblum, which involves exposing monkeys to drug-maternal deprivation combinations in order to "model" human panic disorder.²⁰ Like other maternal deprivation projects, this research is fundamentally flawed.

DESCRIPTION OF PRIOR RESEARCH
Since monkey researcher Leonard Rosenblum assumed directorship of the Primate Behavior Laboratory at the State University of New York (SUNY) in 1963, he has been conducting maternal deprivation experiments with bonnet and pigtail macaque monkeys.²⁰ Prior to this position, he trained under Harry Harlow at the University of Wisconsin, who pioneered maternal deprivation research.^20,21
Early maternal deprivation monkey researchers, including Rosenblum, linked maternal deprivation in monkey infants with the subsequent development of "depression." One of Rosenblum's first studies at SUNY found that pigtail macaque infants were severely disturbed after removal from their mothers. Their loud screams and "massive struggle(s)" showed them to be "distressed" for the entire day, and most of them were deeply "depressed" the next day as well: "Each infant sat hunched over, almost rolled into a ball. . . Movement was rare. . . The movement that did occur appeared to be in slow motion. . . The infant rarely responded to a social invitation or made a social gesture, and play virtually ceased. . ."²² The researchers identified stages of protest and despair in the pigtail macaques similar to findings in rhesus macaques, and called these stages "agitation" and "depression." Unlike rhesus macaques, however, the pigtails showed a third gradual and incomplete phase, which they called "recovery." Mothers and infants showed a resurgence of interaction when reunited.^22,23
When these studies were first performed on bonnet macaques, Rosenblum found that the "depression" phase was absent.^24,25 Further manipulations did include a "depressed" state which was less severe than that seen in other monkeys.26 Gathering food from a foraging device had a therapeutic effect in those monkeys "disturbed" by partial isolation. The success of this "therapy" depended on the monkeys' status in the dominance hierarchy.²⁷
Later, Rosenblum resumed studies on pigtail macaques, finding that they appeared "depressed" during the first separation night, but not during daytime observations over the next few days. Rosenblum viewed the infants' behavior both as a response to the loss of mother and as an attempt to cope without the mother.²⁸

SIGNIFICANCE OF PRIOR RESEARCH
The generalization of these findings to humans--presumably the purpose of maternal deprivation experiments--poses quite a challenge. First, what does it mean to diagnose "depression" in a monkey? Diagnosing "depression" in a monkey undermines the successful ongoing process of clarifying psychiatric diagnoses by using DSM criteria.²⁹ Strict adherence to DSM-IV criteria of depression does not allow for a diagnosis of depression in monkeys, for the subjective experience that includes depressed mood, diminished interest, pleasure, concentration and energy, appetite change, fatigue, feelings of worthlessness and excessive guilt, indecisiveness, and thoughts of death²⁹ cannot be ascertained, but rather only conjectured via inference from monkey behavior. Although clinical psychiatry has become sufficiently sophisticated to appreciate that mood and behavior are neither identical nor predictably derived from one another, animal research allows for no such distinction.
Noted monkey maternal deprivation researcher (and Harlow student) Stephen Suomi, acknowledging dissimilar sensitivities to experiences of early separation in different non-human primate species,³ has become more circumspect in 1995--carefully referring to the condition that resembles depression in monkeys as "something equivalent to depression"⁴ --than he was 24 years earlier, when, like other maternal deprivation monkey experimenters, he referred to the monkey condition as "depression."⁵ Suomi's current appellation, however, is a hedge, for it successfully sidesteps the question by not presuming that the monkeys suffer from the human syndrome of depression, while at the same time implying that, whatever they actually do suffer from, it can serve as a model of human depression. Such a presumption is, in actuality, only an unscientific assumption, supported by neither convincing data nor DSM-IV criteria. Unlike Suomi's more careful, if still inaccurate, description of the monkeys' state, Rosenblum continues to refer to maternally deprived monkeys as "depressed."^22,23,30
Just as there are no valid animal models for schizophrenia, aggression disorders, addiction disorders, and Alzheimer's Disease,^31,32 human depression is a distinctly human disorder for which no animal model exists, regardless of the repeated attempts to model it through maternal deprivation in monkeys.^5,30 Likewise, learned helplessness in monkeys,³³ crowding-induced aggression in rats,³⁴ smoking in mice,³⁵ and alcohol consumption in dogs³⁶ were once, but are no longer, considered animal models of human depression, aggression, and addiction.
One of the difficulties in modelling human behavior through monkey behavior is that the latter varies significantly among different species, as can be seen in "depression-like," social status, aggressive, and child-care behaviors. With respect to "depression-like" behavior in monkeys, Rosenblum himself, as previously mentioned, has shown that it varies among similar monkey species (rhesus, bonnet, and pigtail macaques.)^22-28
Regarding social status factors, Rosenblum has shown that bonnet monkeys' status in the group hierarchy was a key determinant of their success in using effective foraging to reverse partial isolation-induced disturbed behavior.²⁷ However, since hierarchy characteristics vary among monkey species, it is difficult to know the relationship between the positions of monkeys' in their hierarchies and humans in society. For example writer Deborah Blum summarizes:

Squirrel monkeys are a fiercely feminist society. In the wild, the females hang together--the inner circle--and the males hover at the edges, permitted in only during mating season. . .Rhesus macaques live within a rigid and intolerant caste system that has less to do with sex than with the family one is born into. Male and female is not the issue here, it's the monkey version of feudal society. . . Baboons are a patriarchal society dominated by males and fascinated by food. Hunting for the daily meal is one of their favorite occupations...³⁷

Jeffrey Masson supports the notion of questioning the significance of hierarchy by pointing out that both lemurs and mountain pigmy-possums also show female dominance, and that the entire idea of dominance may be fraudulent:

In recent years the idea of the dominant hierarchy has become more controversial, with some scientists asking if such hierarchies are real or a product of human expectations. . .The idea of observing animals engaged in mysterious behavior and charting a tidy hierarchy that produces testable predictions has great appeal for scientists.³⁸

Apart from dominance-hierarchy issues, studies show distinctions between monkey species in other behaviors, i.e., aggression and caretaker. Whereas members of many monkey species often compete aggressively for status, food, and mates, with third parties participating in fights by joining a friend or relative, Tonkean macaques from the Indonesian island of Sulawesi often break up fights among their neighbors.39 Also, in tropical climates males only rarely carry their young, whereas in temperate climates this behavior is common.40 It is apparent that behaviors of one species of monkey do not automatically generalize to another, let alone to humans. This being the case, how can it be determined which monkey data would apply to humans? Which monkey species should be used to generalize its particular "depression-like" behavior, social structure, aggressive habits, or child-care characteristics to humans?
While Rosenblum considers bonnet macaques to be "particularly suited for the experimental study of anxiety problems" because they are gregarious and stable and show low baseline levels of anxiety,²⁰ it is not at all clear that these characteristics make them suitable scientific models for human anxiety studies. More likely, Rosenblum chose bonnet macaques because of "our 30 years of experience with them"²⁰ a rationale that echoes Suomi's observation that animal models are used for practical and pragmatic rather than scientific reasons: "The primary rationale for creating most animal models lies not so much in any obvious impressive strengths of such models as it lies in the problems inherent in conducting research with humans as subjects."⁶ Similarly, Rosenblum has written: "Obvious ethical and practical problems preclude controlled prospective studies with humans, but a range of prospective manipulations are possible with nonhuman primates."²⁰ Given the considerable difficulty in determining which species and which experimental manipulation most clearly resemble human depression, it is hardly surprising that researchers have had difficulty in choosing a model on a scientific basis and have rather relied on non-scientific factors such as availability, expedience, convenience, and personal experience.

continued