Animal Models of Depression—Cruel and Invalid
O’Neil and Moore end their comprehensive review of depression tests conducted on animals by concluding that “it may be time to finally abandon the term ‘model’ with its implication of a direct imitation of the whole syndrome of depression, as this may simply demand too much of what are either tests for antidepressant drug action or tests for a response to stress.”11 This statement casts serious doubt on the validity of many of the methods that are used to study depression in animals, including the practice of inducing “learned helplessness,” the forced-swim test, and others.
Learned Helplessness
Learned helplessness is the traditional measure of depression in animal research, and it continues to be used today. The method involves exposing animals—traditionally dogs or rodents—to electric shocks while blocking any means of escape. Typically, the animals are put in a box with an electrified floor that will not cease delivering shocks no matter what the animals do. Animals are considered to be depressed when they stop their futile attempts to escape the shock and then continue to accept the pain even when escape is offered, such as when a platform that is not electrified is added. O’Neil and Moore explain, “The learned helplessness theory of depression is based on the hypothesis that depression is a learned response to uncontrollable stress in the environment ….”12 They cast doubt on the basic logic behind the learned helplessness theory, noting, “It is also clear from the human literature that not everyone exposed to stressors becomes helpless ….”13
Bourin et al offer up a scathing review of the learned helplessness theory, stating, “Over the years, this model has proven to be fairly limited, for [it] exhibits poor reliability across laboratories and does not parallel the clinical setting with regard to antidepressant treatment, especially concerning SSRIs …. Furthermore, the LH model is not selective to antidepressant (AD) drugs, as anxiolytics have been found to reverse the behavioral outcome ….”14 Selective serotonin reuptake inhibitors (SSRIs) are the most popular and effective pharmaceutical treatments for depression ever developed and sell under brand names such as Prozac (fluoxetine), Zoloft (sertraline), and Paxil (paroxetine). Animals who are given these drugs behave no differently when they receive physical shocks, indicating that the learned helplessness test is truly useless for application to humans.
In discussing the failings of learned helplessness tests, O’Neil and Moore point out, “This test is not routine in many countries where animal experimentation is strictly controlled either by ethical or regulatory supervision or both, as in the case of the UK. The shock regime … is classified as severe and not acceptable for ethical reasons.” And they conclude, “It could be argued that the learned helplessness model induces quite severe levels of suffering for little additional construct validity and no improved predictive validity [in testing pharmaceutical agents].”15 In other words, the learned helplessness test is both extremely cruel and utterly useless. Although the use of this technique is declining, it is often replaced by another cruel research method, the forced-swim test.
Forced-Swim Test
The most popular alternative to inducing learned helplessness is the forced-swim test (FST), in which animals are placed in a container of water with no possible escape. The less the animal tries to escape—by either trying to climb the walls or swimming vigorously—the more depressed the animal is considered to be. This resulting state is referred to as “behavioral despair,” and its use is based on the same flawed principles as the learned helplessness theory.
The problems with “behavioral despair” tests are many, but it is particularly telling that the tests show different results based upon the breed of animal selected, the laboratory conducting the tests, and the protocol used. However, even when all these factors are held constant, the results of behavioral despair tests are unreliable. O’Neil and Moore write, “Thus even where the rank orders of strains were similar, non-subjective measuring systems were used, and animals were obtained from the same suppliers, the different laboratories can report hugely different baseline levels of activity.” The fact that there are basic differences between individual animals is often cited as a cause of these variations in results, and O’Neil and Moore add, “Factors such as housing within each laboratory can have an impact on performance in these tests.”16 They go on to explain that, unbelievably, “The mouse FST model exhibits relatively low validity but is widely used in screening antidepressants (ADS) due to its simplicity.”17
A major flaw in drug tests using animals is that these tests typically look for effects to appear immediately after a single injection of the drug, whereas antidepressants take weeks to show their effects in patients. O’Neil and Moore argue that “one can question if the effect in the single acute test bears any relation to that observed with chronic treatment ….”18 This is a primary reason why animal data have failed to predict clinical outcomes.
The forced-swim test has poor validity and continually gives incorrect information about the effects of pharmaceuticals. The test suggested that psychostimulant compounds, such as amphetamines, would treat depression, but such compounds were ineffective in human patients. Even more damning than these “false-positive” results is the “widely described poor sensitivity of the test for some antidepressants, most notably compounds from the SSRI class ….”19 As with the learned helplessness test, the SSRI data deal a fatal blow to the validity of the forced-swim test.
It is worth noting that another popular behavioral despair test—“tail-hanging”—suffers from the same problem as that of the forced-swim test. In tail-hanging protocols, rodents are suspended upside down, and researchers measure how much time they spend struggling to free themselves. The longer they hang motionless—in “behavioral despair”—the more depressed they are said to be. The fact that these models do nothing to help depressed human patients makes their use nothing short of sadistic.
Chronic Mild Stress Tests
As an alternative to inducing stress through acute pain and fear, researchers also produce chronic mild stress (CMS) in rodents by subjecting them to long-term irritants. This typically includes tilting animals’ cages, keeping their bedding wet, or disrupting their light and dark cycles over a period of days or weeks. Researchers then measure decreases in the animals’ consumption of a sweet solution because this is said to mimic anhedonia, or the inability to experience pleasure, which is a major symptom of depression.
Data from chronic mild stress tests have been so unreliable that they have created a useless and contradictory set of literature. Bourin et al write, “Certain tests, such as the sucrose test in the CMS model, are extremely sensitive to any environmental influences so that it is difficult to obtain stable and repeatable results not only across laboratories, but also within the same laboratory …. Furthermore, in addition to the problem of replicating complex methodological procedures, interpretation of results is problematic when measures are indirect and subjective.”20
O’Neil and Moore note, “A complex procedure such as that involved in the chronic mild stress model almost ensures that every laboratory will have different interpretations of the protocol. Direct replication of the original findings has proved to be beyond many investigators and alternative explanations for the apparent reduction in palatable solutions have been proffered ….” Moreover, when considering the one significant effort ever made to validate this model, O’Neil and Moore note, “Their conclusions were not encouraging. [They] found sucrose consumption to vary across rat strains following stress … and not to be stable within rat strains over time.”21 Test results that cannot be reproduced are not considered scientifically valid.
Social Stress Tests
Ethologists studying animals in the wild have long documented the fact that social relations can cause stress for animals. Most famously, Jane Goodall showed us the trials and tribulations of dominance hierarchies and power struggles among the chimps of Gombe. However, instead of learning from these studies of animals in nature, some researchers decided that social stress was another phenomenon that could be examined in laboratory experiments. Animal researchers have used maternal separation, individual isolation, confinement of groups to small spaces, separation of bonded animals, and other forms of trauma to induce stress in animals.
These animal tests have yet to produce results relevant to the study of depression in humans. Professors O’Neil and Moore explain that “in spite of the emphasis on the effects of stress on numerous markers, it is not clear that what has emerged is a model of depression, but a model of chronic stress per se or a model that detects the effects of anti-panic agents.”22
Animal Research Is Misleading >>
11. O’Neil and Moore 251.
12. O’Neil and Moore 245.
13. O’Neil and Moore 245.
14. Bourin, Fiocco, and Clenet 10.
15. O’Neil and Moore 246.
16. O’Neil and Moore 247.
17. Bourin, Fiocco, and Clenet 10.
18. O’Neil and Moore 247.
19. O’Neil and Moore 246.
20. Bourin, Fiocco, and Clenet 18.
21. O’Neil and Moore 248.
22. O’Neil and Moore 249.
