Newsletter
Volume 30 | 2007 | Number 1
Behavior Analysis and the Best of the Best: A Review of The Cambridge Handbook of Expertise and Expert Performance
By Dr. Thomas S. Critchfield, ABA International President
One of the first people I met upon attending the ABA convention about 25 years ago was Thom McKenzie of San Diego State University, a behavior analyst interested in sport and health/fitness promotion. At the time, Thom was the psychological consultant to the United States National Women’s Volleyball team, and I soon had the opportunity to watch the team practice. My first reaction was simply to be in awe—only by seeing elite athletes up close can one appreciate how different they are from the rest of us. Over time, however, this experience has acquired deeper meaning, for only rarely does one notice behavior analysts working, as Thom McKenzie did, to get the most out of the “best of the best.”
We love to say that behavior analysis will save the world. To do that, however, behavior analysts must make a difference at the highest levels of human accomplishment. To be sure, this happens sometimes. For instance, organizational behavior managers work with some of the biggest and most important corporations. On the whole, however, as Bill Heward of Ohio State University suggested recently(1), there appears to be a negative correlation between the ability level of human beings and the number of applied behavior analytic studies focusing on them.
Would that some guidance existed about how to undertake a systematic investigation of elite human performance! As it happens, a great deal has been written about expertise and methods for studying it, a literature that is nicely summarized in The Cambridge Handbook of Expertise and Expert Performance (Ericsson, Charness, Feltovich, & Hoffman, 2006). The volume is essential reading for anyone who believes in the capacity of behavior analysis to make a difference in high-level human affairs. To be sure, this is not a behavior-analytic treatise, and challenges exist for those who imagine transporting behavior analysis into domains of peak expertise (more on this shortly). An important point not to be lost in the meantime is that that studies of expertise converge on a perspective that should hearten functional thinkers everywhere.
“Gifted” through Experience
A recommended point of departure is Chapter 38, “The influence of experience and deliberate practice on the development of superior expert performance,” by K. Anders Ericsson, who will address the ABA Convention in San Diego as the 2007 Presidential Scholar. Although lay persons – and many psychologists – attribute exceptional performance to innate talents (note how the word “gifted” implies a biological advantage), Ericsson argues for a distinctly environmental perspective, for two reasons. First, contrary to popular assumptions, research on peak performers does not show reliable evidence that they were exceptional at a very early age. Rather, most such individuals appear to have developed their expertise gradually (see Ericsson, Nandagopal, & Roring, 2005). Second, the factor most reliably correlated with this development is a very specific kind of experience called deliberate practice.
To oversimplify, deliberate practice means high-quality engagement with an activity, with frequent feedback, guided by a well-structured “curriculum” designed to minimize errors and extend ability just beyond its current level. One take-home message of the relevant research is that about 10 years of immersion in deliberate practice is necessary to the development of expertise. In other words, those who achieve the most also practice the most. For example, one study cited by Ericsson et al. (2005) found that by early adulthood “gifted” musicians had, by virtue of starting earlier and/or practicing more, accumulated up to twice as much deliberate practice as highly skilled, but not exceptional, peers.
Where is Behavior Analysis in All of This?
The astute reader may recognize in deliberate practice a prescription that is tantalizingly reminiscent of Skinner’s (1968) in Technology of Teaching. Nevertheless, one striking feature of the Handbook is a complete lack of reference (in 42 chapters and 901 pages) to the research and interventions of applied behavior analysis. Not mentioned, for instance, are Hart and Risley’s (1995) elegant analysis of vocabulary development in young children, or the muscular effects of behavioral techniques on academic skill acquisition in the educational disadvantaged (e.g., Johnston & Layng, 1992). This reflects not so much a bias against behavioral psychology as the problem that Bill Heward identified: Applied behavior analysts often focus on instating rudimentary skills, while the Handbook addresses development of the highest levels of expertise. Although behavior analysts tend to be confident of a parallel between these two levels of accomplishment, the sad fact is that few data exist (in the behavior analysis literature) to support such a parallel.
To put the problem into context, consider that, according to one traditional view, there are three kinds of intervention in human services: reducing the complications associated with severe problems (tertiary prevention), reducing the number of minor problems that could get worse (secondary prevention), and preventing the development of problems in people who are functioning adequately but at risk for getting worse (primary prevention). Now consider a continuum of behavior (Figure 1) ranging from that which is severely disordered or impoverished to that which is far better than normal. The three kinds of interventions just mentioned (white boxes at bottom) fall in the center or the left tail of the continuum. The right tail is ignored. Yet interventions are possible to promote behavioral enhancement rather than remediation (gray boxes). Some, as in teaching sport skills to interested youth, generate modest skill levels and potentially apply to large swaths of the population (we might call this primary enhancement). Others, as in gifted education, harness the potential of a relatively few individuals who already are performing better than the norm (secondary enhancement). Yet others, as in the grooming of concert pianists, seek to create the most exceptional of all repertoires (tertiary enhancement). It is to this last mission that the Handbook points—although, as I’ll note in concluding, it points there somewhat hesitantly.
Figure 1. A continuum of behavior and its relation to several kinds of interventions.
Challenges in Establishing a Behavior Analytic Approach to Expertise
For those interested in a behavior analytic approach to exceptional performance, the Handbook provides an invaluable perspective on how others have explored this topic. It contains detailed descriptions of what expert performers do differently than others, how they acquire their skills, and how these things manifest in a variety of skill domains (including chess, medicine, mathematics, and software design). The most useful, and challenging, chapters for behavior analysts, however, are those that explain the scientific methods that have been used to study expertise.
Behavioral research often adopts a simple-to-complex strategy in which stripped-down repertoires are instated and understood, via experimental methods and direct observation, in great depth before levels of complexity are added. Much research on expertise, by contrast, has attempted to deconstruct existing high-level repertoires using descriptive methods. Thus, for example, interview techniques may be employed to reconstruct a learning history, and think-aloud protocols may be employed to determine what experts are thinking during overt performances. Neither method is standard operating procedure in behavior analysis, and both are likely to raise concerns among methodological purists.
In the interest of avoiding knee-jerk reactions, however, both cases deserve close inspection. Self-reports prompted by interviews are potentially fallible data sources, but they are also a form of behavior, and behavior may or may not be under control of a given event. Thus self-reports are not necessarily flawed, and under the right circumstances they correlate closely with the events they purport to describe. For example, one well-validated procedure has been used to track alcohol consumption by problem drinkers over periods of months or years (see Critchfield, Tucker, & Vuchinich, 1998).
Think-aloud protocols are another kind of self-report – trickier because they describe events taking place inside the skin that cannot be corroborated as easily as an overt act like drinking alcohol. Yet these protocols have been put to good use in several behavior analytic investigations (Austin, 2000). Once again, the key is to match the method to circumstances that are likely to generate useful data (e.g., Critchfield & Epting, 1998).
As an aside, one reason think-aloud protocols are appealing is that deliberate practice is thought to promote the development of covert precurrent behaviors that, consistent with Skinner’s (1966) analysis of problem solving, promote more effective overt behavior (Austin, 2000). For instance, in a study that required subjects to practice digit span (remembering sequences of numbers) extensively over many weeks (Ericsson, Chase, & Faloon, 1980), one subject who initially could recall sequences only a few digits long learned to reliably master sequences of up to 80 digits. He did this, in part, by thinking of short sequences as a single time for running the mile (e.g., 3:57.21 instead of 3, 5, 5, 2, 1). This spontaneous application of the strategy known as chunking would not have been obvious without some insight into the subject’s silent thoughts.
If self-report-based methods of data collection are distrusted, what are the alternatives for analyzing personal histories and covert responses? Perhaps, for instance, behavior analysts can employ experimental methods to build superior performance from scratch. In a sense, every behavior analytic laboratory study requiring steady-state performance is a study of expertise, so behavior analysts have some relevant tools to apply here. That said, the responses under study, especially in basic behavioral research, rarely resemble the rich, nuanced skill sets of expert performers in the everyday world. Moreover, for practical reasons, experiments rarely can reproduce the extensive histories that research suggests contribute to expertise. Overall, it may be impossible to study expertise without a combination of methods (indeed, many of the Handbook’s generalizations are based on a convergence of experimental and descriptive findings). In the end, behavior analysts who are wed exclusively to experiments and direct observation must decide whether methods will guide the topic of investigation, or the other way around (e.g., Critchfield, Haley, Sabo, Colbert, & Macropoulis, 2003).
Room at the Top
Expertise is highly valued in society, and there are many domains of expert skill. Moreover, as Ericsson (Chapter 1) notes, rapid advances in knowledge and technology dictate that “the life-long quest for improved adaptation to task demands will not be limited to experts anymore” (p. 17). No wonder that the study of expertise is a burgeoning scientific movement, and how sad that behavior analysis has contributed so little to it.
A few decades ago, behavior analysts might have been able to set the agenda for the scientific study of expertise, but it is difficult to gain entry into a maturing scientific field. Does a place at this table remain for behavior analysts who wish to study and promote high-level skill? The answer to this question is imbedded in a feature of the literature from which the Handbook samples: there is plenty of analysis, but not much intervention. In other words, the typical developmental trajectory of established experts is well understood, but can we apply this knowledge to reliably create new expertise? Here is an opportunity that is tailor made for behavior analysts, as even some in the expert-performance community understand. For example, Felovich, Pioleta, & Ericsson (Chapter 4) note that research on reinforcement
...Might contribute to our understanding of how some people manage to persevere through the very long periods of practice and experience, involving both successes and inevitably many failures, that we now know are so essential to the development of expert levels of skill. How to scaffold sustained, consistent, purposeful effort, over very long periods of time and despite inevitable setbacks, appears at this time to be one of the great puzzles to be solved in developing a science of human excellence (p. 45)
Thus, although studies of expertise suggest how the best performers got that way, deriving a replicable technology of behavioral enhancement from this knowledge is a work in progress. Behavior analysis should interest those who care about expertise because behavior analysts excel at building repertoires. The Handbook should be interesting to behavior analysts – especially those with designs on making the world a better place – because it defines this opportunity, and raises the compelling question of which repertoires we will choose to build.
(1) Heward spoke at the 2006 American Psychological Association convention in New Orleans upon accepting the 2006 Fred S. Keller Behavioral Education Award from the Division of Behavior Analysis. Congratulations, Bill!
References
Austin, J. (2000). Performance analysis and performance diagnostics. In J. Austin & J. E. Carr (Eds.), Handbook of applied behavior analysis (pp. 321-349). Reno, NV: Context Press.
Critchfield, T. S., & Epting, L. K. (1998). The trouble with babies and the value of bath water: Complexities in the use of verbal reports as data. Analysis of Verbal Behavior, 15, 65-74.
Critchfield, T. S., Tucker, J. A., & Vuchinich, R. E. (1998). Self-report methods. In K. A. Lattal and M. Perone (Eds.), Handbook of methods for the experimental analysis of human behavior (pp. 435-470). New York: Plenum.
Ericsson, K. A., Charness, N., Feltovich, P. J ., & Hoffman, R. R. (Eds.) (2006). The Cambridge Handbook of Expertise and Expert Performance. New York: Cambridge University Press.
Ericsson, K. A., Chase, W. G., & Faloon, S. (1980). Acquisition of a memory skill. Science, 208, 1181-1182.
Ericsson, K. A., Nadogapal, K., & Roring, R. W. (2005). Giftedness from the expert-performance perspective. Journal for the Education of the Gifted, 28, 287-311.
Hart, B., & Risley, T. R. (1995). Meaningful differences in the everyday experience of young American children. Baltimore: Brookes.
Johnson, K. R. & Layng, T. V. J. (1992). Breaking the structuralist barrier: Literacy and numeracy with fluency. American Psychologist, 47, 1475-1490.
Skinner, B. F. (1966). An operant analysis of problem-solving. In B. Kleinmuntz (Ed.), Problem solving: Research method and theory (pp. 225-257). New York: Wiley.
Skinner, B. F. (1968). The technology of teaching. New York: Appleton-Century-Crofts.