A Review of Spatial Ability Research
Total Page:16
File Type:pdf, Size:1020Kb
View metadata, citation and similar papers at core.ac.uk brought to you by CORE provided by Engineering Design Graphic Journal (ASEE - American Society for Engineering Education) spring 2008 A Review of Spatial Ability Research James L. Mohler Purdue University Abstract Spatial ability research has been approached from several psychological vantages since its beginnings in the late 1800s. Th is contribution attempts a summation of spatial ability research, beginning with a historical vignette and a major section on each psychological approach including the psychometric, developmental, diff erential and information pro- cessing approach. Of importance is what each approach has contributed to our knowledge of spatial ability. ___________________________________________________________________________________ WHY IS SUCH A Th e chronology of spatial ability research can REVIEW NEEDED? be broken into four major periods of activity. Table 1 shows this author’s chronology and the In the fall of 1996, Miller provided an ex- associated themes or approach. While an assort- cellent historical review of spatial visualization ment of sources provide in-depth historical ac- research. His article chronicled the various ap- counts (Carroll, 1993; Eliot & Smith, 1983), proaches to spatial visualization literature docu- a brief historical vignette seems appropriate to mented in past issues of the Engineering Design begin this contribution; setting the stage for a Graphics Journal. Miller’s article is a well-known review of the major research traditions and their (and well-referenced) starting point for thesis or contributions. dissertation research projects on the topic. How- ever, there is much literature outside the bounds Table 1 of the Engineering Design Graphics Journal. To complement Miller’s work, this contribution de- Chronology of Research with Themes and Approach tails spatial ability research from various branch- Date Range Themes and Approach es of psychology and other fi elds. Essentially, this article documents seminal pieces of literature Acknowledgement of a spatial 1880 - 1940 factor separate from general that, when combined with Miller’s historical re- intelligence through psychomet- view, provide a more holistic view of the fi eld of ric studies spatial ability research. Acknowledgement of multiple 1940 - 1960 space factors through psycho- metric studies; emergence of THE BEGINNINGS OF myriad spatial assessments THE RESEARCH Psychometric studies into 1960 - 1980 cognitive issues; emergence of With implications for nearly every techni- developmental and differential cal fi eld, spatial ability continues to be an active research thread of research found throughout many dis- Effect of technology on ciplines. As early as 1880, Sir Francis Galton re- measurement, examination, 1980 - and improvement; emergence ported on his experimental inquiries into mental of information processing imagery. Since that time, researchers have defi ned research spatial ability in numerous ways, contending over its constituents and creating various methods for measuring it. Mohler - 19 v o l u me 72 n u mb e r 3 A Historical Vignette during this period, knowledge of spatial ability– Although credit belongs to Galton (1880, its development and diff erentiation–blossomed. 1911) as being the initiator of the research, pub- lications with a spatial focus did not emerge un- While prior research themes have continued, til the early 1920s. Contributions from 1880 to from 1980 to today researchers have focused on 1940 acknowledged and defi ned spatial ability the impact of technology on measurement, ex- as separate from general intelligence. Th rough amination, and improvement of spatial ability. In the work of Th orndike (1921), Kelley (1928), El addition, much attention has been turned toward Koussy (1935), and Th urstone (1938), research- understanding spatial ability from the informa- ers regarded spatial ability as a capacity separate tion processing perspective. from the general intelligence factor (g) defi ned by Spearman (1927). From this 100-year history of research one thing remains clear: spatial ability is a set of com- From 1940 to 1960, researchers focused their plex, cognitive abilities about which there are still energies on defi ning what comprised spatial abil- many questions. Each of the research approaches ity, but not without diffi culty. While a few re- provides a unique contribution. Th e following searchers attended to this area, many deemed the sections will delve into these research approaches, ability unimportant. Many viewed spatial func- providing an outline of signifi cant endeavors and tioning as a “lower ability” due to its practical contributions. manifestations. Adding to this undervaluation, confusion within the burgeoning community PSYCHOMETRIC RESEARCH created additional diffi cultly (D’Oliveira, 2004; Lohman, 1979a). Because of diff ering factor One of the initial challenges posed to spatial analysis techniques and the use of diff erent spatial research was distinguishing it from the general in- ability tests, researchers adopted contradictory telligence factor. Two major groups with diff ering names and defi nitions for spatial factors (Cooper views pursued intelligence research. Research in & Mumaw, 1985). Th ey also included confl ict- Britain followed Spearman in focusing on intel- ing numbers of factors (see Hegarty & Waller, ligence as a single factor, whereas research in the 2005). Nevertheless, spatial testing obtained an U.S. viewed intelligence as composed of multiple important foothold due to large-scale assessment factors. Th e former work was pursued by Spear- conducted in the Army Air Forces (Guilford & man (1927), Burt (1949), and Vernon (1950) Lacy, 1947). By the end of this period, research- and the latter work was conducted by Th urstone ers agreed that spatial ability was not unitary and (1950), Cattell (1971), and Guilford (1967). many spatial tests were available (see Eliot & Smith, 1983). Initially researchers had diffi culty distinguish- ing spatial ability factors from intelligence because From 1960 to1980, several divergent ap- several of the spatial factors load quite heavily on proaches to spatial ability research emerged. Wit- general intelligence (spatial visualization tests, for kin (1950) and Gardner’s (1957) psychometric example). Typically intelligence has been viewed studies examined cognitive issues such as learn- hierarchically and taxonometrically, with the for- ing styles. Developmental studies by Piaget and mer emerging fi rst (Gustafsson, 1988). Figure 1 Inhelder (1971) examined how spatial ability shows a basic hierarchical view of the structure of develops through childhood to adulthood. Dif- human abilities and the juxtaposition of spatial ferential researchers focused on areas of diff erence abilities (Smith, 1964). in spatial ability, particularly as it relates to diff er- ences across gender. Work by Maccoby and Jack- As shown in Figure 1, when mental tests are lin (1974) serves as the much-referenced contri- analyzed using factor analysis, the fi rst factor to bution in this area. Due to the varied approaches be extracted typically corresponds to g. Once g is 20 - Engineering Design Graphics Journal spring 2008 Figure 1. Hierarchical structure of human abilities (Smith, 1964). removed, the tests typically fall into two groups: verbal-numerical (v:ed factor) and the spatial- mechanical-practical (k:m factor). If there are enough tests in the battery being used, the two subgroups can be divided further into minor factors, such as verbal, numerical, or spatial and manual. Scientifi c and empirical work that is more re- cent has attempted to defi ne hierarchical models of intelligence and specifi c aspects of those mod- els (Snow & Lohman, 1989; Snow, Kyllonen, & Marshalek, 1984). Due to its extensive inclusion Figure 2. Example of the radex model of datasets, the best-known contemporary factor of intelligence (Guttman, 1954). analytic survey is Carroll (1993). Of importance to this review was Carroll’s dis- intelligence, which demonstrates the positioning cussion of a hierarchical “three-stratum theory” of spatial ability in juxtaposition with verbal and of ability that “could be accommodated within, mathematical ability. or show correspondences with, radex theories Th e three abilities shown in Figure 2 have that assume hierarchical structures” (Carroll, psychological importance and can predict occu- 1993, p. 654). Carroll identifi ed three hierarchi- pational and educational success. While Carroll cal strata (narrow, broad, and general) into which (1993) discussed arguments against this “three- cognitive abilities fell. Radex theories, the earliest stratum theory,” the sheer magnitude of the data of which Carroll credits to Guttman (1954), are and subsequent studies present a compelling ar- typically taxonomic (rather than hierarchical). gument for support of the radex model. Howev- Figure 2 shows an example of the radex model of er, some research acknowledges that hierarchical Mohler - 21 v o l u me 72 n u mb e r 3 and radex models can mesh quite well and even fl uency. Th is theory was the basis for intelligence complement each other (Snow, et. al, 1984). tests that yield a profi le of individual performance from several ability scores, rather than the single The Acknowledgement mark. of a Spatial Factor Th e published identifi cation of spatial abil- Multiple Space Factors ity was a 1921 paper by Th orndike. He drew an Th rough subsequent research and using ab- important distinction among three