Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Methods in Behavioral Toxicology: Current Test Batteries and Need for Development Helena Hanninen Behavioral toxicology is an applied branch of psychology with a very practical aim: to detect neurotoxic effects for the planning of prevention among exposed populations. The earliest research in clinical and epidemiological behavioral toxicology was done within occupational medicine by psychologists working close to it or by medical doctors with some familiarity with psychological methods. Most of the research in the area is still done in this context. It provides the best possibilities for collaboration with the other disciplines working toward the same goal. In this context, however, the connections of behavioral toxicology with other branches of psychology- applied, as well as theoretical have been loose. Re- ports are seldom published in psychological journals but rather in those of occupational medicine or neurotoxicology because of the practical relevance of study results within these disciplines. As a consequence, research on behavioral toxicology has been little known to those working in other fields of psychology. Another consequence is a limitation of scientific feedback from colleagues in other branches of psychology. However, psychologists doing research on behavioral toxicology do use the knowledge and methods developed in other fields of psy- chology. Methods of behavioral toxicology originate from different branches of applied and academic psychology: clinical psychology, neuropsychology, aptitude testing, psychophysiological psychology, and in recent years, cognitive psychology. Tests were borrowed from 39
40 HELENA HANNINEN these fields, mostly without trying to incorporate the corresponding theoretical framework into behavioral toxicology. The theoretical framework has been regarded as less important than the efficiency of the tests in detecting neurotoxic effects and their applicability in epi- demiological research. TEST BATTERIES NOW IN USE Epidemiological and Field Studies The methods now in use in epidemiologically oriented behavioral toxicology compose an interesting variety of psychological tests. Eight studies published in 1985-1987 serve as examples (Table 1~. The variety of tests used reflects the history of behavioral toxicology and, to a certain extent, the advances of psychological testing during that period. One of the first comprehensive test batteries for the detection of neurotoxic effects was designed at the Institute of Occupational Health in Finland, in connection with a study on carbon disulfide (Hanninen, 1971~. It was composed of the best tests on hand at that time. First a very large selection of tests was used; there were tests of clinical psychology and neuropsychology, as well as aptitude tests standard- ized at the institute. Then, after preliminary data analyses, those tests of different areas of functioning were selected which seemed to be sensitive and suitable for use in clinical practice as well as in epide- miological research. Over the years the battery has been modified, and in each of the Finnish studies a somewhat different selection of tests has been used, depending on the exposure agent in question and the overall study design. Study 1 (Mantere et al., 1984) in Table 1 is one example. Descriptions of the tests were published in a sepa- rate booklet (Hanninen and Lindstrom, 1979), and several of them have spread to wider use in behavioral toxicology: the Santa Ana dexterity test, the Bourdon-Wiersma Vigilance test, the Benton test for visual memory, and selected subtasks of the Wechsler Adult Intelligence Scales (WAIS). Tests of WAIS and WMS (Wechsler Memory Scale) as well as the Benton Visual Retention test have been much used also because they are well documented and have been stan- dardized in several countries. Other neuropsychological tests have been adapted to behavioral toxicology, too. Studies 2 and 3 are other examples of test batteries composed mainly of much-used neuropsychological tests. The tests are not the same, however. The battery used in study 2 (Baker et al., 1984) was com- posed of subtasks of the Wechsler Scales; verbal memory was empha-
METHODS IN BEHAVIORAL TOXICOLOGY 41 sized. Santa Ana was used for psychomotor ability. The British battery (Cherry et al., 1984) used in study 3 (Cherry et al., 1985) has three psychomotor tests and three for visual-motor performance. Of the latter, the Visual Searching task was devised by Goldstein as an indicator of brain damage, and Trail Making is from the Halstead Reitan Battery. The two most applied tests of the WAIS, Digit Sym- bol and Block Design, are also included in this battery. Memory is assessed by one test only. The British National Adult Reading Test (NART) is included as a measure of premorbid ability. The battery used in study 4 (leyaratnam et al., 1986) concentrated even more on visual, motor, and visual-motor functions; each was measured with several tests. Additionally there was Digit Span for memory performance. Study 5 (Maizlish et al., 1985) is an example of studies that apply methods developed within experimental psychology. There were no common elements with the five previous test batteries, but the bat- tery covers principally the same broad functional domains as these do. Each of the tests contains a long series of measurements (about 15 minutes each), and gives detailed information about the function · . . In quesnon. Many of the recent test batteries in behavioral toxicology include single computer-assisted tests adapted from other branches of psy- chology. The Neurobehavioral Evaluating system (NES) developed by Baker and Letz (Baker and Letz, 1986) and used in studies 6 (Maizlish et al., 1987) and 7 (Fidler et al., 1987) is totally computerized. Some of its tests can be described as computerized versions of certain con- ventional tests that have shown sensibility in toxicity testing; some were chosen or designed as promising new tests for that purpose. Being an extensive battery, the NES provides an opportunity to choose relevant methods for different kinds of studies. In addition to the tests used in studies 6 and 7, the regular sequence of NES tests also includes a verbal Associate Learning task and a vocabulary test for obtaining a measure of initial intelligence level. The last study in Table 1 (Williamson and Teo, 1986) is another example of new methodological approaches in behavioral toxicology. Its tests were selected on the basis of information-processing theory, which also provides the framework for interpreting the results. BEHAVIORAL TOXICOLOGY IN CLINICAL PRACTICE Behavioral toxicology aims at examining and identifying toxic ef- fects on the functional capacity of the central nervous system (CNS).
42 - Go at - - ct it ~ E i_ ~ a` - Go - - >>I ~ .s; In U be ·_ V, ._ x - o .> o V, · - I - Go - - - Cal ~ - Go - - _' ~ Ct cn a; .s EN ~ ¢ c _ .~ ~ ~ O .~.m ~ = ~ ~ - ~ ~ ~ _ o C 3 0D 5 ~ ~ ~ ~ a 0 0 ~ b4 m ct O - CJ) ~ v, - .= - E~ co ED ~4 ._ ~ ' ~ ~ r i~ ~~ ~ , ~ ~ In ~ _ ~ 5 ~ ~ 2 ~ ' 3 a, ~ ~ lo, At, ~
43 - oo - o En o in ao ~ .m _ ~ Cal ~ oo - . - ._ in ~ i_ oo - - _ N A; ~0 - - a, N ~5 cn ~ ED o ~ ~ ·~ ~ in ~ a' ~ =0 C C_ :1 - ~ u =~> ~ m~ m :^ O ~ .~ ~ c C ~ E ~ -A E so cat in Pa cn cn an cot cr. Z Z Z Z ZZ Z o U. ._ ~ C ~ m. ~ ~ o ~ X ~ ° ~ ~ ~ ~ v: uo cn ~ c~ cn cn cn ZZ Z Z Z Z Z cn O ° :- :- u .5 u° o ~ == ~ == cn 0 cn ~ ~ <~, ~ 3 .s ~ ~ o o.= o o o.= o ~ cn ~ ~ ~ ~ ~ ~ ~ U ~ ~ ~ C~ d. U~ ~ ~ oO b4 ._ - ._ ._ 11 . ~ - CC ~ 1-, o o o o ~ ~ 11 ~ ·` o ~ U, ~o o · V ~ _ ~ {,4 ._ _ 11 ~ ·_ .` ~ ~ rD ._ ~ - ~s 11 o .~. ' o ~ _ ~ ~ o - 11 ~ ° rQ ~ 11 ~ ·` O O 11 ~ ·~ a~ , ·- 11 ~, u, ' I_' ·~ ¢ o u, ·- ttS ~ . E~ ~ _ ·_ 4 - u,
44 HELENA HANNINEN The subject matter links it with neuropsychology. The methodologi- cal choices described above reflect this link, but not consistently. The neuropsychological orientation is particularly relevant when neuro- behavioral methods are used for clinical assessment of individuals afflicted by neurotoxic exposures. In Scandinavia and in some other European countries, psychologi- cal assessment is a rather regular part of diagnostic procedure when a neurotoxic occupational disease is suspected. Moreover, at least in Finland, it is rare for referral of such a patient to be based on psycho- logical examination done in a neurological clinic or a public health center. Neuropsychologists in other countries have been less aware of the neurotoxic syndrome, but the situation seems to be changing rapidly. Neurobehavioral dysfunctions seen in intoxication patients pro- vide useful information that directs the choice of methods when the early effects are studied, and vice versa: results of the epidemiologi- cal studies have guided the selection of tests in clinical test batteries. Table 2 lists the tests included in three diagnostic test batteries: the one used by Bolla-Wilson and Bleecker in the Johns Hopkins School of Medicine (BolIa-Wilson and Bleeker, 1987~; the Pittsburgh Occupa- tional Exposures Test Battery, or POET (Ryan et al., 1987), which is meant for use at clinics as well as in research; and the battery now in use at the Finnish Institute of Occupational Health (IOH). The first battery represents traditional neuropsychological meth- odology; the two others were designed especially for neurotoxic ef- fects. Differences among the three batteries are not striking. All of them sample roughly the same broad functional domains, with one or more tests. The tests selected are not the same, however. Only four WAIS tests appear in all three batteries. Seven tests are in- cluded in two of them. Additionally, each battery contains tests not present in the two others. There are also differences in emphasis. The Finnish battery attempts a detailed description of the motor per- formance and its disturbances, whereas there is only one ordinary psychomotor test in the two other batteries. On the other hand, the set of memory tests included in POET allows a much more elaborate description of the memory function than the Finnish battery does, and the battery used by Bolla-Wilison and Bleecker is the only one that assesses verbal functions that may be affected by toxic exposure. BASES OF TEST SELECTION Published studies in behavioral toxicology seldom give reasons for selection of test methods: reports emphasize results, not methodol-
45 - =< ~ L4~1 ~ of-; ~ I ~ ~ ~ ,y 1, I_ ~ cr: ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ¢ V) o ~5 cr. ._ x o o o 2 ~ y $~Y~s~52~ Tic · ~ Ct A: ~ ~ ~ g ~ 5 at, ¢ Em ~ Hi ~ di to ~ ~ Go V) ._ 11 d. ~ o .` 60 a' 11 ~ Go . ~ ~ id. O ~ O ~ C o o U. ~ A car ~ _ - O ~ · ._ _ 11 11 Cal . ~ ·~ O _ ~ ~ Ct O ~ it; 11 ~ Cal \D . ~ . ~ O ~ . - o ~ _ \_ A; ~~ _ US A u, 11 U) ~0 .. G ~ _ U. ·= - .. o ~ · - u, - 1 o - o · - ~ . ~ co cn ~ ~ u, u, ~ o >` ·~ o o (s ~
46 HELENA HANNINEN ogy. This does not mean that there was no rationale for choosing the particular tests used. Very often, however, the rationale can only be guessed. In the choice of tests, the first questions considered are the foBow- ing: What behavioral/neuropsychological functions are likely to be affected? Which tests are available for examining those functions? The answer to the second question yields, in principle, a wide selec- tion of relevant tests. To choose among them, further questions must be considered. Important criteria are (1) previous empirical evidence of the sensitivity of the tests with regard to the effect looked for; (2) other previous documentation of the test: its psychometric proper- ties, general population norms, documentation of its construct valid- ity, etc.; (3) suitability and acceptability of the tests for the subjects to be tested; and (4) costs and benefits of the tests: costs in terms of time, expertise, and equipment needed; benefits in terms of informa- tion provided by the method. The existing differences in test selection (Tables 1 and 2) reflect different answers to these questions or a different importance given to them. Most test series include measures of a wide area of behavioral functions. This indicates that the authors either expected the effects to be widespread and diffuse, or did not have specific expectations concerning the functions to be affected. It is also possible that they had such expectations, but wanted to sample a broader area of func- tion to be better able to circumscribe the effect. Some differences in the emphasis given to different functional areas were also found: some batteries emphasized psychomotor tasks; others, cognitive tests, or memory tests in particular. These differences seemed not to be related to the exposure agent under study, but rather to reflect differ- ent concepts of the nature of the neurotoxic syndromes in general. The tests chosen to measure different behavioral areas varied more than the general structure of the batteries. Some classic neuropsychological tests seem to be rather well-established general tools in behavioral toxicology, but there were also batteries that had very few common elements with the others, or none at all. One explanation is that the authors represented different branches of psychological research and preferred tests that best corresponded to their education and experi- ence. Moreover, the studies were conducted in different countries where different tests may have been available and well documented with general population norms, etc. It seems also that the previously established sensitivity of the test in regard to toxic effects was not considered equally important by all of researchers. Rather, the authors can be classified as conservative
METHODS IN BEHAVIORAL TOXICOLOGY 47 or progressive. The former prefer the "old" tests because there is previous documentation about their sensitivity and the possibility of comparing their results with earlier ones. At the same time, the users of these tests may be troubled by their weaknesses: question- able reliability, low acceptability by the subjects, laborious scoring, etc. The "progressive" authors, on the other hand, strive to introduce recent theoretical or methodological progress in different branches of psychology into the field of behavioral toxicology. Their problem is that in using new promising and potentially sensitive methods, they lose the possibility of comparing their results with early research. Also, in case of negative results, they have difficulty in deciding whether the exposure in question actually had no behavioral effect or whether the promising methods indeed were not sensitive enough to display them. Consideration of the population to be investigated is also neces- sary when selecting a test. The suitability of a test depends partly on the general educational level and culture of the subjects, and on the cultural and educational homogeneity of the subject group. In cul- turally and linguistically very heterogeneous populations the meth- odological choices are limited mainly to sensory and motor tests. Moreover, the practical circumstances of the study outlined by the study design, the testing facilities (particularly in field studies), and economic matters may strongly restrict methodological choices, mak- ing a wise test selection a real challenge. THE NEUROBEHAVIORAL CORE TEST BATTERY The Neurobehavioral Core Test Battery (NCTB) was designed by an expert group of the World Health Organization (WHO). It is in- tended for use in health hazard evaluations and field studies when the testing time is limited and the use of sophisticated equipment is not possible Johnson, 1987~. The seven tests included in the NCTB can also be recommended as common tests in more comprehensive test batteries, to allow comparison of results obtained by different research groups. The NCTB includes Pursuit Aiming for motor steadiness, Simple Reaction time for attention and response speed, Digit Symbol for perceptual motor speed, Santa Ana for manual dexterity, the Benton Visual Retention test for visual perception and memory, and the Digit Span test for auditory memory. The Profile of Mood States (McNair et al., 1981) is included as a measure of affect. When tests were chosen for the NCTB, special consideration was given to their appli- cability for examining working populations all over the world where
48 HELENA HANNINEN neurotoxic hazards exist. The tests had to be minimally dependent on the culture and education of the subjects. It is evident that such a short series of tests cannot sample all the possible effects of all possible neurotoxicants. Moreover, it may not be sensitive enough to discover subtle effects of low-level exposure. Nevertheless, it covers the functions that are most likely to be af- fected and can be expected to be able to detect neurotoxic effects in more hazardous exposure situations. FUTURE DIRECTIONS: SOME CRUCIAL ISSUES Neuropsychological Description/Definition of Neurotoxic Syndromets) Selection of relevant methods would be easier if we knew more about the CNS dysfunction being studied. Is there a general neuro- toxic syndrome, affecting many different areas of behavioral function? If so, then the effect is most likely to be seen in those functions that are most vulnerable in CNS dysfunctions in general, or in functions for which the most accurate and reliable tests exist. If many performances are only slightly impaired, mere chance can decide which of the tests yields a result that reaches the desired level of statistical significance. If there is a general diffuse neurotoxic syndrome (at the behavioral level), is it similar to that seen in other diffuse brain conditions, or does it have any unique, or at least semispecific, features? In spite of the advances of neurotoxicology during the last decades, not too much is known today about the mechanisms that mediate between the toxic effect on the brain and behavioral dysfunction. Several effects on brain tissue have been found, these effects being somewhat different for different exposure agents. Nevertheless, the operation of differ- ent mechanisms may produce similar effects on behavioral output. The toxic syndrome differs from other diffuse brain syndromes, and the syndromes caused by different toxicants differ from each other, mainly as far as the effect on the brain is selective and some areas of the brain are affected more than others. Very little is known about this today. There appear to be both a general diffuse behavioral dysfunction caused by all neurotoxic exposures, or most of them, and more spe- cific impairments that may be agent dependent and related to the different sensitivity of various brain structures to the agent in question. It can be hoped that advances in neurotoxicology research will yield more information about the mechanisms and sites of action of various neurotoxicants. In addition, better acquaintance with the advances
METHODS IN BEHAVIORAL TOXICOLOGY 49 of neuropsychology might help to put the recent and future pieces of information together, and to formulate fruitful hypotheses concern- ing the neuropsychological nature of the toxic syndromes, to guide further research. Even without knowing the pathological processes in the brain that are behind the syndromets), their more accurate descriptions should be attempted. This can be done, for example, by investigating crucial behavioral disturbances with theoretically well-based test combina- tions that allow a more detailed analysis of these disturbances. This approach was used in the study by Williamson and Teo (1986; study 8 in Table 1), and in a study on solvent-induced memory impairment by Stollery and Flindt (1988~. Natural History of the Neurotoxic Effects The time course of behavioral effects is also poorly known. The severity of an effect is apparently a function of both the intensity and the duration of exposure, as demonstrated by Mutti and coworkers (1984) in their study on styrene. However, it can also be assumed that the time course and the dose-effect curve differ for various ef- fects. The different exposure history of the subjects may, in fact, explain many of the discrepancies in empirical results. Longer duration or a higher level of exposure can be expected to cause not only more severe but also more widespread behavioral effects. At the other end of the exposure history, accentuation of behav- ioral impairment depends on the time elapsed since cessation of ex- posure. Clinical observations on patients with occupational intoxication and the few follow-up studies reported so far indicate different reversibility of the affected functions: intellectual functions and visual memory tend to recover, whereas deficits in verbal memory and motor performance are more often irreversible or progressive (Lindstrom et al., 1982; Orbaek and Lindgren, 1988~. Different recovery curves for varying dysfunctions suggest a different brain pathology behind them. Dilemma Between Conservative and Progressive Approaches There are several challenges for new and more sophisticated meth- ods in this area of research: the need to increase the sensitivity of tests, the need to get a better understanding of the phenomena being studied, and the need to increase the efficiency of research, particularly in large-scale field studies, by using timesaving computer-based methods. The problem is that the application of many new methods will
50 HELENA HANNINEN increase the methodological diversity of the field, complicate the comparison of results, and slow down the accumulation of a common body of knowledge. Another serious problem is that new methods with unknown sensitivity to neurotoxic effects should not be used as the only behavioral methods in studies evaluating the behavioral tox- icity in exposed populations. For both practical and ethical reasons there is a very limited possibility of conducting purely methodologi- cal research on exposed populations. Thus, it can be expected and even recommended that the methods which have been most useful in previous studies will in the future also be used in behavioral toxicity studies, complemented when possible with promising new methods. However, experimentation with new methodological approaches in clinical situations and in field studies, whenever possible, is vitally important too. The construction of good new tests is an arduous procedure. Even more work is usually needed to gather different types of validity data for new methods. Empirical research done with a test in differ- ent populations and addressing different questions increases the va- lidity of the interpretation of its results. Recognizing this fact, the authors of the NES battery organized international collaboration to gather versatile data on the reliability and validity of the NES tests. A similar procedure is going on concerning the WHO battery, even though its tests have already been widely used. This is, I think, the right way to proceed with new methodological approaches. Further Questions of Test Selection Simple or Complex Tasks ? Some behavioral toxicologists prefer testing simple elemental functions (sensory, motor, and cognitive) instead of complex ones, partly be- cause this strategy may yield a more precise description of the dysfunction. Another reason for measuring narrow elemental functions is that they offer fewer possibilities for compensating an impaired function by use of other, unaffected ones. Simple reaction time is one example of a well-circumscribed elemental function sensitive to neurotoxic expo- sures. However, two other arguments speak for the use of more complex tasks. Although requiring complex interaction of several brain struc- tures, such tasks may be sensitive to effects not detected by tests focusing on narrow functions. In addition, if many elemental func- tions are impaired, then performance may be more grossly impaired
METHODS IN BEHAVIORAL TOXICOLOGY 51 in a test requiring several of them than in tests measuring single functions. The optimal strategy would be to use tests of varying complexity, but in a real-life research situation this may be impossible, because a too long test battery tends to distort the study design by raising the costs and increasing the risk of dropouts, thereby decreasing the sta- tistical power of the data analyses. Reliability of Tests and Stability of Functions High reliability (i.e., minimizing measurement error) is one of the basic qualifications of a good test. Reliability has several compo- nents: objectivity of the measurement, internal consistency of the test, and stability of the test result over time. The latter requirement is somewhat problematic. Changes in the test score over time do not reflect only the error variance, but also the instability of the measured function (i.e., its sensitivity to situational factors). Certainly, there is no sense in measuring very unstable functions. On the other hand, stressing-the importance of test-retest reliability too much may lead to a selection of insensitive tests. According to the results of a recent solvent study (Hanninen et al., 1986), those tests that displayed the best dose-response relationship were least stable in the exposed group. It seems that variability of the performance level over time, and even during a single exposure session, is itself a characteristic of subjects handicapped by a neuro- toxic effect. Changes in Personality, Mood, and! Affect The behavioral effects of neurotoxic substances include changes in mood, affect reaction, and personality. These changes can be present alone or along with performance decrements (Cranmer and Goldberg, 1986~. Even in the latter case they often are the most disturbing and handicapping effects from the point of view of afflicted subjects and their closest environments. Compared to the work done to find good methods for the sensory, motor, and cognitive dysfunctions, research on the changes in affect has been badly neglected. Test batteries described in Tables 1 and 2 did not contain measures of affect or personality, except for the POMS (Profile of Mood States) included in the NIBS and the WHO battery. However, most studies contained symptom enquiry with questions on mood states, emo- tional distress, fatigability, etc. In some studies, as well as clinical
52 HELENA HANNINEN examinations, various personality tests have been applied (Forzi et al., 1975; Lindstrom, 1984; Maroni et al., 1977~. Valid assessment of emotional changes is difficult. Moods and emotions have a large interindividual and intraindividual variation, their quantification is problematic, and confounding factors are diffi- cult to rule out. One problem is the difficulty in distinguishing primary toxic effects from more secondary reactions due to diminished func- tional capacity or to anxiety about being exposed to a toxic substance. Though the role played by exposure in emotional changes and other psychological symptoms is difficult to evaluate, their assessment should not be neglected when toxic effects are studied. Improvement and refinement of methods to be used for this purpose are major challenges for behavioral toxicologists. Other Areas for Method Development i, , Deficient memory and learning ability are common complaints among subjects exposed to neurotoxic substances. In some studies the tests of learning and memory were among the best detectors (Eskelinen et al., 1986; Hanninen et al., 1976), but in several others they yielded more marginal or negative results. Even patients with marked sub- jective memory deficits sometimes have normal or near-normal scores in memory tests. Improvement of the sensitivity of these tests is another major challenge. Sensitivity of some memory tests can be improved by improving their psychometric qualities. For example, the Digit Span of the Re- vised WAIS yields more accurate scores than the old Digit Span test, and sensitivity of the WMS Associate Learning can be improved by increasing the number of difficult items. Yet improving old tests may not be enough. Memory is not one single function but has several components. Selection of the best tests to detect neurotoxic impairment of memory and learning requires a detailed description of the deficit in question, i.e., the application of paradigms of contemporary research on memory and its impairments. Further development is similarly needed in regard to assessment of attention and its disturbance in neurotoxic conditions. In this area, the main difficulty may be finding or devel- oping tests that yield a valid measurement of attention (and not of sensory or motor speed or something else) and are still short and simple enough to be applicable in epidemiologic research. Another area needing special consideration is the category of defi- cits in the executive functions: goal formulation, planning, carrying
METHODS IN BEHAVIORAL TOXICOLOGY 53 out activities, etc. (Lezak, 1984~. Though it may be difficult to find good methods to apply these disturbances as research tools, their assessment should not be neglected when examining patients with suspected neurotoxic diseases. SUMMARY The variety of tests in use has been illustrated by eight studies published in the last four years. In addition, three clinical test batter- ies have been described. These examples were selected to give a picture of the primary methodological approaches, and not to present a comprehensive review. The most usual approach has been an eclectic use of tests pre- viously employed in other areas of psychology, mainly in neuro- psychological research and practice. Many of these tests have dis- played good sensitivity to neurotoxic effects. The variety of choices is large, but some of the tests have been used by many research groups in several countries. Some of the recent studies have striven toward the application of newer methodological and theoretical approaches to behavioral toxi- cology. Certain tests have been computerized to increase the objec- tivity and cost-effectiveness of the testing, and to obtain a more detailed measure of the performance and its deficits. The theoretical para- digms of contemporary psychology have been applied to acquire a better and more elaborated understanding of the neuropsychological dysfunction being studied. However, the new methods often require equipment or expertise that is not always available when research on neurotoxicity is ur- gently needed, or they are still lacking sufficient validation with regard to their sensitivity to neurotoxic substances. The use of older tests with documented validity is therefore to be recommended in many field study situations. At the same time, however, it is important to work on new methodological paradigms aiming particularly at a better neuropsychological description of the neurotoxic syndrome and its agent-specific variants. Future research should also be directed to the time course of neurotoxic effects during continuing exposure and after its cessation. Special consideration should be given to the toxic-induced changes of mood, affect, and personality, as well as to the deficits of atten- tion, learning, and memory. These are common concomitants of a neurotoxic affliction for which practical, reliable, and sensitive methods are still lacking.
54 HELENA HANNINEN REFERENCES Baker, E. L., and R. Letz. 1986. Neurobehavioral testing in monitoring hazardous workplace explosures. J. O. M. 28:987-990. Baker, E. L., R. G. Feldman, R. A. White, J. P. Harley, C. A. Niles, G. E. Dinse, and C. B. Berkeley. 1984. Occupational lead neurotoxicity: A behavioral and electrophysi- ological evaluation. Study design and one year results. Br. J. Ind. Med. 41:a352-361. Bolla-Wilson, K., and M. L. Bleeker. 1987. Neuropsychological impairment following inorganic arsenic exposure. J. O. M. 29:500-503. Cherry, N., H. Hutchins, T. Pace, and H. A. Waldron. 1985. Neurobehavioral effects of repeated occupational exposure to toluene and paint solvents. Br. J. Ind. Med. 42:291-300. Cherry, N., H. Venables, and H. A. Waldron. 1984. Description of the tests in the London School of Hygiene test battery. Scand. J. Work Environ. Health 10(Suppl 1):1 19. Cranmer, J. M., and L. Goldberg. eds. 1986. Proceedings of the workshop on Neurobehavioral effects of solvents. Neurotoxicology 7:45-56. Eskelinen, L., M. Luisto, L. Tenkanen, and O. Mattel. 1986. Neuropsychological meth- ods in the differentation of organic solvent intoxication from certain neurological conditions. J. Clin. Exp. Neuropsychology 8:239-256. Fidler, A. T., E. L. Baker, and R. E. Letz. 1987. Neurobehavioral effects of occupa- tional exposure to organic solvents among construction painters. Br. J. Ind. Med. 44:292-308. Forzi, M., M. G. Cassitto, R. Gilioli, G. Armeli, and V. Foal 1975. Personlichskeit- fehlentwicklungen in Arbeitern bei der electolytischen Chlor-Alkali-Gewinnung. Pps. 7~73 in Proceedings of the 2nd Industrial and Environmental Neurology Congress, E. Klimkova-Deutschova and E. Lukas, eds. Prague: Universitas Karlova. Hanninen, H. 1971. Psychological picture of manifest and latent carbon disulphide poisoning. Br. J. Ind. Med. 28:37~381. Hanninen, H., L. Eskelinen, K. Husman, and M. Nurminen. 1976. Behavioral effects of long-term exposure to a mixture of organic solvents. Scand. J. Work Environ. Health 2:240~255. Hanninen, H., and K. Lindstrom. 1979. Behavioral test battery for toxico-psychologi- cal studies used at the Institute of Occupational Health in Helsinki. Institute of Occupational Health Reviews 2:1-58. Hanninen, H., E. Tuominen, K. Rantala, and R. Luukkonen. 1986. Psychological tests in the detection of early neurotoxic effects: Construction and preliminary valida- tion of a screening test battery. Tyoterveyslaitoksen tutkimuksia 4:68-71 (Finnish with English summary). Jeyaratnam, J., K. W. Boey, C. N. Ong, C. B. Chia, and W. O. Phoon. 1986. Neuropsychological studies on lead workers in Sinzanore. Br. J. Ind. Med. 43:62 629. -o--r - Johnson, B., ed. 1987. Prevention of Neurotoxic Illness in Working Populations. New York: John Wiley & Sons. 257 pp. Lezak, M. 1984. Neuropsychological assessment in behavioral toxicology Developing techniques and interpretative issues. Scand J. Work Environ. Health 10(Suppl. 1):25- 29. Lindstrom, K., M. Anffl-Poika, S. Tola, and A. Hyytiainen. 1982. Psychological progno- sis of diagnosed chronic organic solvent intoxication. Neurobehav. Toxicol. Teratol- ogy 4:581-588.
METHODS IN BEHAVIORAL TOXICOLOGY 55 Lindstrom, K. 1984. The Rorschach test in behavioral toxicology. Scand. J. Work Environ. Health 10(Suppl. 1):20-23. Maizlish, N. A., G. D. Langolf, L. W. Whitehead, L. J. Fine, J. W. Albers, J. Goldberg, and P. Smith. 1985. Behavioural evaluation of workers exposed to mixtures of organic solvents. Br. J. Ind. Med. 42:579-590. Maizlish, N., M. Schenker, C. Weisskopf, J. Seiber, and S. Samuels. 1987. A behavioral evaluation of pest control workers with short-term, low-level exposure to the oganophosphate diazinon. Am. J. Ind. Med. 12:153-172. Mantere, P., H. Hanninen, S. Hemberg, and R. Luukkonen. 1984. A prospective follow-up study on psychological effects in workers exposed to low level of lead. Scand. J. Work Environ. Health 10:43-55. Maroni, M., C. Bulgheroni, M. G. Cassitto, F. Merluzzi, R. Gilioli, and V. A. Foal 1977. A clinical, neurophysiological and behavioral study of female workers exposed to 1,1,1-trichlorethane. Scand. J. Work Environ. Health 3:16-22. McNair, D. M., M. Lorr, and L. F. Droppelman. eds. 1981. Manual for the Profile of Mood States. California: Education and Industrial Testing Services. 29 pp. Mutti, A., A. Mazzucchi, P. Rustichelli, G. Frigeri, G. Artini, and I. Franchini. 1984. Exposure-effect and exposure-response relationships between occupational expo- sure to styrene and neuropsychological functions. Am. J. Ind. Med. 5:275-281. Orbaek, P., and M. Lindgren. 1988. Prospective clinical and psychometric investiga- tion of patients with chronic toxic enchephalopathy induced by solvents. Scand. J. Work Environ. Health 14:37~4. Ryan, C. M., L. A. Morrow, E. J. Bromet, D. K. Parkinson. 1987. Assessment of neuropsychological dysfunction in the workplace: Normative data from the Pitts- burgh Occupational Exposures Test Battery. J. Clin. Exp. Neuropsychology 9:665- 679. Stollery, B. T., and M. L. Flindt. 1988. Memory sequelae of solvent intoxication. Scand. J. Work Environ. Health 14:4508. Williamson, A. M., and R. K. Teo. 1986. Neurobehavioural effects of occupational exposure to lead. Br. J. Ind. Med. 43:37~380.
