Are Animal Models Relevant in Modern Psychiatry?

Figure

The use of animals in medical research became firmly established in 1865, with the publication of An Introduction to the Study of Experimental Medicine by Claude Bernard.¹ This scientific discourse laid the groundwork for the study of comparative physiology between animals and humans. Half a century later, the animal model was introduced into the behavioral sciences by early theorists Pavlov (classical conditioning), Watson (behaviorism), and Skinner (operant conditioning). Later, the animal model was used to investigate conditions ranging from maternal deprivation to depression and learned helplessness.

Most early psychiatric drugs were discovered through serendipity rather than through the use of animal models. Isoniazid, originally used to treat tuberculosis, was found to possess mood-altering properties and was marketed as the first antidepressant in 1957.^2,3 MAO inhibitors originated from an effort to develop antituberculosis medications; they were superseded by tricyclic antidepressants (TCAs), which were discovered by clinical observation.

The potential psychotropic effect of chlorpromazine, originally used as an anesthetic adjunct in a Paris hospital in 1952, was discovered by a military surgeon later in the same year.^4,5 Thus, the phenothiazines came from a search for better pre-anesthetic agents.

The Australian physician John Cade⁶ reported the calming effect of lithium in humans in 1949. The first benzodiazepine, chlordiazepoxide (Librium), was discovered accidentally in 1955.⁷ The first studies of benzodiazepines were unsuccessful attempts to treat patients with schizophrenia.⁸

In contrast to discoveries made through chance observation, newer psychotropic drugs, such as SSRIs, were discovered through the process of rational drug design. Five SSRIs (citalopram, fluvoxamine, fluoxetine, paroxetine, sertraline) were produced independently by 5 different companies.⁹ Rational drug design remains the main driving force behind the development of modern psychiatric drugs.

Animals as model systems in psychiatry

Since the mid-20th century, researchers have designed animal models of stress, anxiety, depression, and obsessive-compulsive conditions in the laboratory to develop, test, and validate drugs to treat human disorders.^10-13 Rats and mice are most commonly used in specific behavioral tests, such as the despair test, tail suspension test, and open field test.

Current animal models of human psychiatric conditions face the same methodological limitations as they did 30 years ago. According to Beach¹⁴:

The validity of interspecific generalization can never exceed the reliability of intraspecific analysis; and the latter is an indispensable antecedent of the former. . . . Significant comparison of a particular type of behavior in two different species is impossible unless and until the behavior has been adequately analyzed in each species by itself.

This hypothesis is conditional on the existence (or availability) of animal models to accurately mimic human psychiatric conditions. In reality, the overwhelming majority of mental disorders recognized by DSM, the International Statistical Classification of Diseases and Related Health Problems (published by the World Health Organization), and the American Psychiatric Association do not have a counterpart in laboratory animals. For those human conditions that are considered to have animal homologues, there often exist critical causal mechanisms that differ between humans and animals, which raise methodological questions about the soundness and relevance of these animal models.¹⁵

Because of the multifactorial nature of conditions such as depression and anxiety and the ambiguities inherent in psychiatric diagnosis and treatment, the use of animal models in psychiatry presents unique challenges-unlike those found in other medical disciplines. In most cases, animal models represent a compromise because the cause and mechanism of the human condition under investigation may not be fully understood. In addition, researchers are using a relatively simple system (receptor activation or inactivation) to represent a more complex and less readily studied system (human mental disorders). While examples can be found to demonstrate common and conserved modes of action of neurotransmitter chemicals throughout phylogenetically remote organisms, this approach has its limits when studying complex systems, such as the human CNS.¹⁶ According to molecular biologist Marc van Regenmortel¹⁷:

The reductionist method of dissecting biological systems into their constituent parts has been effective in explaining the chemical basis of numerous living processes. However, many biologists now realize that this approach has reached its limit. Biological systems are extremely complex and have emergent properties that cannot be explained, or even predicted, by studying their individual parts. The reductionist approach-although successful in the early days of molecular biology-underestimates this complexity and therefore has an increasingly detrimental influence on many areas of biomedical research, including drug discovery and vaccine development.

Animal models, in general, have not been subjected to the rigors of evidence-based medicine. Few systematic reviews or meta-analyses have been conducted to compare treatment outcomes in laboratory animals with outcomes in clinical trials. Overall, the animal model has performed poorly as a predictive modality of human outcome in these reviews.^18-22

Examples of animal models in psychiatry

Behavioral despair. The behavioral despair model is commonly used to screen candidate antidepressants. Like other TCAs, imipramine was screened using the Porsolt forced swim test. In this test, the animal (a rat or a mouse) is placed in a container with cold water and is forced to swim until exhausted; it is then briefly taken out of the water. This is repeated until the animal reaches a state of helplessness and stops swimming. Although immobility time is reduced by antidepressant agents such as imipramine, significant strain differences have been reported.²³

The Porsolt forced swim test has been criticized on the grounds that the state of helplessness is more a strategy of survival than a sense of “despair.” The increased immobility simply demonstrates a positive behavioral adaptation, in which the animal has learned that it cannot escape and is conserving energy until it is removed from the water. In addition, while the test is reported to distinguish antidepressants and neuroleptics from anxiolytics, false positives have been reported for a number of other compounds, including stimulants, convulsants, anticholinergics, pentobarbital, and opiates.^24,25

Canine acral lick dermatitis. This is considered by some researchers to be a suitable animal model for the study of obsessive-compulsive disorder (OCD) in humans.²⁶ However, most cases of acral lick dermatitis in animals have an underlying allergic cause; if the allergen is eliminated, the condition resolves. In humans, OCD is an anxiety disorder characterized by intrusive thoughts, unrelated to allergies. Although SSRI treatment may be effective for both acral lick dermatitis in animals and OCD in humans because of a shared neurotransmitter response, this does not confer homology on the animal model. According to Nonneman and Woodruff,²⁷ “If every aspect is fully isomorphic between the animal model and the human condition, including cause and mechanism, the model is homologous.” Clearly, the OCD dog model does not fulfill these criteria. In addition, the dog model ignores the existence of a genetic component and the presence of comorbidities, which are thought to play an important role in human OCD.^28,29

Transgenic mice and the study of anxiety-like endophenotypes. Humans and mice share approximately 97% of their working DNA and approximately 24,000 genes per body cell.³⁰ In light of these facts, scientists have turned their attention to the study of differential gene expression patterns in genetically altered mice to investigate anxiety through the corticotropin-releasing hormone system, the serotonin system, and the GABA system.³¹ However, because genes work in networks and animals are examples of complex systems, small changes at the gene level can have major consequences for the individual. Thus, it is irrelevant to point to observed similarities in genetic makeup (including transgenes) between species, since the details of the differences are in the interactions between conserved genes, not in the genes themselves.³²

As Hirst and colleagues³³ have shown in the example of the serotonin system, it is unlikely that serotonin type 6 knockout mice will be useful for validating the serotonin type 6 receptor as a therapeutic target because of unexpected species differences in both receptor regional distribution and pharmacological profiles. It is now evident that slight variations in rodent and human amino acid sequences can lead to unexpectedly large differences in the pharmacology of the receptors, with potentially disastrous effects for drug development. What has not been clearly documented until the study by Hirst and colleagues is that mouse receptors could be significantly different from rat receptors.

Role of non-human primates in psychiatric and neurological research. The use of non-human primates by Harlow and associates³⁴ to demonstrate the effects of maternal deprivation has been well documented. In experiments conducted between 1957 and 1963, they removed baby rhesus monkeys from their mothers and observed the effects of partial and total social isolation. Some of the monkeys were kept in solitary confinement for up to15 years. These studies have been criticized on both ethical and methodological grounds.¹⁵

Diagnosing “depression” in a monkey is at odds with the successful ongoing process of clarifying psychiatric diagnoses by using DSM criteria. It is not possible to ascertain feelings of worthlessness and excessive guilt, indecisiveness, and thoughts of death from observations of monkeys.³⁵ Harlow’s student Stephen Suomi became more circumspect in 1995, when he carefully referred to the condition in monkeys as “something equivalent to depression” rather than “depression.”

The non-human primate is considered by some researchers to be the most appropriate model for the study of brain function. Among non-human primates, the rhesus macaque is the animal of choice for cognitive studies. While there may be similarities between the brains of humans and non-human primates, the monkey brain is not a scaled down version of the human brain.³⁶ Rather, each primate brain is the unique result of evolutionary biology, molded over millions of years in response to environmental, social, and genetic influences (Figure). With the human brain, the effects of cultural evolution are also considered.

There are numerous differences in the anatomy and physiology of the CNS in monkeys and humans, including differences in locations of specialized areas in the brain. The visual 1 area accounts for 10% of the total cortex in the monkey but only 3% in humans, and anatomically corresponding visual areas in monkeys and humans can perform very different functions.³⁷ The human brain’s architecture and physiology is far more complex than that of the monkey brain. One indication of this is the length of time it takes for the brain to develop in its major phase: 136 days for monkeys and 470 days for humans.³⁸ Other significant differences include the number of synapses a human neuron makes (between 7000 and 10,000) compared with the number a rhesus monkey neuron makes (between 2000 and 6000) and the expression of at least 91 genes involved in a variety of neural mechanisms that differ between monkeys and humans.^37,39 According to Kreiman and associates³⁷:

Even though the hippocampus appears to be one of the most conserved areas of the brain (most similar among mammals), there are still considerable differences. Neurotransmitter receptor distribution varies widely between species. For example, there is an additional small layer of high-density kainate receptors in the deepest part of the hippocampal molecular layer in the monkey, but not in humans. The inhibitory GABA_A receptors are located with high density in the human CA 1 hippocampal region, but not in the same region in monkeys. These results demonstrate considerable changes of the regional and laminar distribution of important signaling molecules in an otherwise evolutionary conservative brain region.

The validity of the animal model in regulatory toxicology

From a regulatory perspective, psychiatric drugs are required to undergo animal testing to demonstrate safety and efficacy.⁴⁰ Safety relates to toxicity testing, which requires a rodent species (usually a rat) and a non-rodent species (usually a dog) to undergo single-dose (acute), repeated-dose, and long-term exposure to a candidate compound. The acute toxic dose (LD50) is the median lethal dose of a compound required to kill 50% of a group of test animals. Repeated-dose exposures usually require 14- to 28-day studies, while long-term exposure requires up to 90 days in the rat and up to 12 months in the dog. Significant variations in the toxicity profile of a test compound may occur because of species differences in absorption, distribution, metabolism, and excretion (ADME).

Human lethal overdose figures are compiled by national poisons information centers. These are empirical data obtained from human accidental or deliberate poisoning and overdose (A. Vale, personal communication, 2009). Human lethal overdose cannot be predicted from animal LD50 values (Table). Animal studies are in fact poor predictors of human outcome with respect to drug toxicity.^41-44The use of retrospective correlation is often inadvertently confused with prediction.

A classic study by Olson and colleagues⁴⁵ looked at the concordance between adverse events seen in humans and data obtained from preclinical animal toxicity studies. While this study measured sensitivity, it ignored specificity; therefore, its conclusions are largely irrelevant to the great prediction debate.³² Although this study has been quoted in support of animal use as a predictive modality, Olson⁴⁵ stated that “this study did not attempt to assess the predictability of preclinical experimental data to humans.”

In addition to the challenges presented by differences in ADME between species, other variables can influence the outcome of animal studies. The laboratory environment is inherently stressful for the animals. In addition to caged housing, which provides little or no environmental enrichment, even routine procedures can lead to significant changes in physiological parameters correlated with stress (eg, serum or plasma concentrations of corticosterone or glucose, heart rate, and blood pressure).⁴⁶ While these changes may not be considered to be of much significance with respect to routine drug testing, the same cannot be said of drugs specifically designed to treat stress and similar conditions. According to Zinberg and Robertson,⁴⁷ the difference between an animal’s natural setting and an artificial laboratory environment cannot be underestimated; it may produce major pharmacological effects. Variables such as group housing versus single housing, type of bedding, light-dark cycle, and handling by humans can affect the outcome of a study.^48-50

Evidence-based psychiatry

Evidence-based medicine aims to apply the best available evidence gained from the scientific method to clinical decision making. Clinical research is the solid ground of medicine, whereas biological theory is a necessary but changing superstructure.⁵¹ In an era of the human genome and advanced noninvasive screening techniques, it is perhaps timely to examine the relevance of the animal model in modern psychiatry. This becomes all the more interesting when we consider the notion that psychiatry is the only medical discipline that attempts to treat a conceptual abstraction-namely the human mind-in addition to organs and physiological processes.¹⁵ Although molecular psychiatry provides some insight into the mechanism of mental illness, external psychosocial factors that influence human behavior fall well outside of its scope.⁵²

Some researchers continue to justify the need for invasive cognitive studies in animals (particularly the non-human primate) on grounds that single neuron studies are not feasible in humans.^53,54 That is no longer the case, since single neuron activity can now be studied in patients with Parkinson disease who undergo deep brain stimulation and in patients with epilepsy in whom electrode probes are used to identify epileptic foci. We need to reconcile ourselves with the fact that the level of detail that has been achieved in the study of the macaque brain will likely never be achieved in studies of the human brain.⁵⁵

Current trends indicate that diagnosis and treatment of psychiatric disorders in the 21st century will rely more on an integrative approach, including the best of genomic advances (eg, human brain transcriptome; pharmacogenomics in drug development geared toward personalized medicine); noninvasive screening techniques combined with ethical pharmacological studies, such as pharmaco-magnetoencephalography^56,57; and clinical observation and other relevant human-based methodologies.

Empathy in psychiatry

There is an urgent need to acknowledge the role of empathy in science. The inclusion of empathy as a fundamental tenet is especially important in psychiatry and psychology, as an antidote to the mechanistic view of disease and the undervaluing or even ignoring of psychological and spiritual components.¹⁵ Animal researchers have traditionally been circumspect about including any notion of empathy in their work, for fear of being stigmatized as anthropomorphic.

In contrast, some researchers have openly expressed the moral conflict they face when using non-human primates in experiments.⁵⁸ These researchers occasionally acknowledge the very negative effects of family separation, isolation, and boredom, in addition to the pain and suffering inflicted on the animals during the experiments.

Several studies have indicated that mammalian species, especially the chimpanzee, have a sense of self and self-awareness.^59,60 This has led some scientists, particularly behaviorists, to suggest that these animals should be afforded special protection from pain, suffering, and incarceration.⁶¹ In addition, studies of other mammalian species, including mice and rats, suggest that they also possess awareness of self and even more subtle “human” qualities associated with empathy and social joy.⁶² All of this empirical evidence points to a need to embrace a paradigm of “ultimate concern in science.”⁶³

References:

1. Bernard C. An Introduction to the Study of Experimental Medicine. Greene HC, trans. New York: Macmillan & Co, Ltd; 1927.
2. Waterstradt K. A transitory psychosis occurring twice after isoniazid therapy [in German]. Dtsch Med Wochenschr. 1957;82:1138.
3. Jackson SL. Psychosis due to isoniazid. Br Med J. 1957;2:743-746.
4. Charpentier P, Gailliot P, Jacob R, et al. Recherches sur les dimé-thylaminopropyl-N phénothiazines substituées. Comptes rendus de l’Académie des sciences (Paris). 1952;235:59-60.
5. Laborit H, Huguenard P, Alluaume R. Un noveau stabilisateur végétatif (le 4560 RP). Presse Med. 1952;60:206-2088.
6. Cade JF. Lithium salts in the treatment of psychotic excitement. Med J Aust. 1949;2:349-352.
7. Sternbach LH. The discovery of Librium. Agents Actions. 1972;4:193.
8. Preskorn SH. CNS drug development. Part I: the early period of CNS drugs. J Psychiatr Pract. 2010;16:334-339.
9. Preskorn SH. CNS drug development. Part II: advances from the 1960s to the 1990s. J Psychiatr Pract. 2010;16:413-415.
10. Senay EC. Toward an animal model of depression: a study of separation behavior in dogs. J Psychiatr Res. 1966;4:65-71.
11. McKinney WT Jr, Bunney WE Jr. Animal model of depression. I. Review of evidence: implications for research. Arch Gen Psychiatry. 1969;21:240-248.
12. Dinsmoor JA, Bonbright JC Jr, Lilie DR. A controlled comparison of drug effects on escape from conditioned aversive stimulation (“anxiety”) and from continuous shock.Psychopharmacologia. 1971;22:323-332.
13. Bliss EL, Zwanziger J. Brain amines and emotional stress. J Psychiatr Res. 1966;4:189-198.
14. Beach FA. Animal models for human sexuality. Ciba Found Symp. 1978;(62):113-143.
15. Cohen M. A critique of maternal deprivation experiments on primates. http://www.mrmcmed.org/mom.html. Accessed December 19, 2011.
16. Garcia-Reyero N, Habib T, Pirooznia M, et al. Conserved toxic responses across divergent phylogenetic lineages: a meta-analysis of the neurotoxic effects of RDX among multiple species using toxicogenomics. Ecotoxicology. 2011;20:580-594.
17. Van Regenmortel MH. Reductionism and complexity in molecular biology. Scientists now have the tools to unravel biological and overcome limitations of reductionism. EMBO Rep. 200;45:1016-1020.
18. Knight, A. Systematic reviews of animal experiments demonstrate poor contributions toward human healthcare. Rev Recent Clin Trials. 2008;3:89-96.
19. Knight A, Bailey J, Balcombe J. Animal carcinogenicity studies: implications for the REACH system. Altern Lab Anim. 2006;34(suppl 1):139-147.
20. Lindl T, Voelkel M, Kolar R. Animal experiments in biomedical research. An evaluation of the clinical relevance of approved animal experimental projects [in German]. ALTEX. 2005;22:143-151.
21. Perel P, Roberts I, Sena E, et al. Comparison of treatment effects between animal experiments and clinical trials: systematic review. BMJ. 2006;334:197.
22. Pound P, Ebrahim S, Sandercock P, et al; Reviewing Animal Trials Systematically (RATS) Group. Where is the evidence that animal research benefits humans? BMJ. 2004;328:514-517.
23. Porsolt RD, Bertin A, Jalfre M. “Behavioural despair” in rats and mice: strain differences and the effects of imipramine. Eur J Pharmacol. 1978;51:291-294.
24. Borsini F, Meli A. Is the forced swimming test a suitable model for revealing antidepressant activity? Psychopharmacology (Berl). 1988;94:147-160.
25. Porsolt RD, Bertin A, Jalfre M. Behavioral despair in mice: a primary screening test for antidepressants. Arch Int Pharmacodyn Ther. 1977;229:327-336.
26. Stein DJ. An animal model of obsessive-compulsive disorder. Arch Gen Psychiatry. 1992;48:517-521.
27. Nonneman AJ, Woodruff ML, eds. Animal models and the implications of their use. Toxin-Induced Models of Neurological Disorders. New York: Springer; 1994.
28. Rasmussen SA. Genetic studies of obsessive compulsive disorder. In: Hollander E, Zohar J, Marazziti D, Oliver B, eds. Current Insights in Obsessive Compulsive Disorder. Chichester, England: John Wiley & Sons; 1994:105-114.
29. Mineka S, Watson D, Clark LA. Comorbidity of anxiety and unipolar mood disorders. Annu Rev Psychol. 1998;49:377-412.
30. Church DM, Goodstadt L, Hillier LW, et al; Mouse Genome Sequence Consortium. Lineage-specific biology revealed by a finished genome assembly of the mouse. PLoS Biol. 2009;7:e1000112. doi:10.1371/journal.pbio.1000112.
31. Bakshi VP, Kalin NH. Animal models and endophenotypes of anxiety and stress disorders. In: Davis KL, Charney D, Coyle JT, Nemeroff C, eds. Neuropsychopharmacology. The Fifth Generation of Progress. New York: Raven Press/American College of Neuropsychopharmacology; 2002:883- 900.
32. Shanks N, Greek R. Animal Models in Light of Evolution. Boca Raton, FL: Brown Walker; 2009.
33. Hirst WD, Abrahamsen B, Blaney FE, et al. Differences in the central nervous system distribution and pharmacology of the mouse 5-hydroxytryptamine-6 receptor compared with rat and human receptors investigated by radioligand binding, site-directed mutagenesis, and molecular modeling. Mol Pharmacol. 2003;64:1295-1308.
34. Harlow HF, Dodsworth RO, Harlow MK. Total social isolation in monkeys. Proc Natl Acad Sci U S A. 1965;54:90-97.
35. Psychiatric Disorders. Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV). AllPsych Online. http://allpsych.com/disorders/dsm/html. Accessed May 4, 2011.
36. Crick F, Jones E. Backwardness of human neuroanatomy. Nature. 1993;361:109-110.
37. Kreiman G, Fried I, Koch C. Single-neuron correlates of subjective vision in the human medial temporal lobe. Proc Natl Acad Sci U S A. 2002;99:8378-8383.
38. de Graaf-Peters VB, Hadders-Algra M. Ontogeny of the human central nervous system: what is happening when? Early Hum Dev. 2006;82:257-266.
39. Kreiman G. Single unit approaches to human vision and memory. Curr Opin Neurobiol. 2007;17:471-475.
40. International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use. Development Safety Update Report. August 17, 2010.
41. Shanks N, Greek R, Greek J. Are animal models predictive for humans? Philos Ethics Humanit Med. 2009;4:2.
42. Suter K. What can be learned from case studies? The company approach. In: Lumley C, Walker S, eds. Animal Toxicity Studies: Their Relevance for Man. Lancaster, England: Quay Publishing; 1990:71-78.
43. Fletcher AP. Drug safety tests and subsequent clinical experience. J R Soc Med. 1978;71:693-696.
44. Lumley C. Clinical toxicity: could it have been predicted? Pre-marketing experience. In: Lumley C, Walker S, eds. Animal Toxicity Studies: Their Relevance for Man. Lancaster, England: Quay Publishing; 1990:49-56.
45. Olson H, Betton G, Robinson D, et al. Concordance of the toxicity of pharmaceuticals in humans and in animals. Regul Toxicol Pharmacol. 2000;32:56-67.
46. Balcombe JP, Barnard ND, Sandusky C. Laboratory routines cause animal stress. Contemp Top Lab Anim Sci. 2004;43:42-51.
47. Zinberg NE, Robertson JA. Drugs and the Public. New York: Simon and Schuster; 1972.
48. Hurst JL, West RS. Taming anxiety in laboratory mice. Nat Methods. 2010;7:825-826.
49. Longordo F, Fan J, Steimer T, et al. Do mice habituate to “gentle handling?” A comparison of resting behavior, corticosterone levels and synaptic function in handled and undisturbed C57BL/6J mice. Sleep. 2011;34:679-681.
50. Nakayasu T, Kato K. Is full physical contact necessary for buffering effects of pair housing on social stress in rats? Behav Processes. 2011;86:230-235.
51. Ghaemi SN. A Clinician’s Guide to Statistics and Epidemiology in Mental Health: Measuring Truth and Uncertainty. New York: Cambridge University Press; 2009.
52. Engel GL. The need for a new medical model: a challenge for biomedicine. Science. 1977;196:129-136.
53. Grinvald A, Hildesheim R. VSDI: a new era in functional imaging of cortical dynamics. Nat Rev Neurosci. 2004;5:874-885.
54. Janssen P, Srivastava S, Ombelet S, Orban GA. Coding of shape and position in macaque lateral intraparietal area. J Neurosci. 2008;28:6679-6690.
55. Van Essen DC, Lewis JW, Drury HA, et al. Mapping visual cortex in monkeys and humans using surface-based atlases. Vision Res. 2001;41:1359-1378.
56. Oldham MC, Konopka G, Iwamoto K, et al. Functional organization of the transcriptome in human brain.Nat Neurosci. 2008;11:1271-1282.
57. Hall SD, Barnes GR, Furlong PL, et al. Neuronal network pharmacodynamics of GABAergic modulation in the human cortex determined using pharmaco-magnetoencephalography. Hum Brain Mapp. 2010;31:581-594.
58. Barnes D. The use of nonhuman animals in psychobiological and behavioral research. In: Natelson NB, Cohen MJ, eds. In: Proceedings from Future Medical Research Without the Use of Animals: Facing the Challenge; May 15-16, 1990; Tel Aviv, Israel.
59. Gallup GG Jr. Chimpanzees: self-recognition. Science. 1970;167: 86-87.
60. Kaneko T, Tomonaga M. The perception of self-agency in chimpanzees (Pan troglodytes). Proc Biol Sci. 2011 May 4; [Epub ahead of print].
61. Goodall J. A plea for the chimps. New York Times Magazine. May 17, 1987:108-110.
62. Panksepp J. Neuroevolutionary sources of laughter and social joy: modeling primal human laughter in laboratory rats.Behav Brain Res. 2007;182:231-244.
63. Panksepp J. Toward a science of ultimate concern. Conscious Cogn. 2005;14:22-29.