Much new is occurring regarding the domain of Gv. Below is a new review of the Gv research and a proposed heuristic framework. This is then followed by select excerpts from our (Schneider and McGrew, 2018) upcoming CHC update chapter in the CIA book, where we add some caution regarding new “proposed”Gv frameworks.
A Heuristic Framework of Spatial Ability: A Review and Synthesis of Spatial Factor Literature to Support its Translation into STEM Education. Article link.
Jeffrey Buckley & Niall Seery & Donal Canty
An abundance of empirical evidence exists identifying a significant correlation between spatial ability and educational performance particularly in science, technology, engineering and mathematics (STEM). Despite this evidence, a causal explanation has yet to be identified. Pertinent research illustrates that spatial ability can be developed and that doing so has positive educational effects. However, contention exists within the relevant literature concerning the explicit definition for spatial ability. There is therefore a need to define spatial ability relative to empirical evidence which in this circumstance relates to its factor structure. Substantial empirical evidence supports the existence of unique spatial factors not represented in modern frameworks. Further understanding such factors can support the development of educational interventions to increase their efficacy and related effects in STEM education. It may also lead to the identification of why spatial ability has such a significant impact on STEM educational achievement as examining more factors in practice can help in deducing which are most important. In light of this, a synthesis of the spatial factors offered within existing frameworks with those suggested within contempo-rary studies is presented to guide further investigation and the translation of spatial ability research to further enhance learning in STEM education.
Keywords Spatial ability . Spatial factors . STEM education . Human intelligence
Click on image to enlarge.
The following are select sections of our Gv chapter in the forthcoming CIA book.
Visual processing (Gv) can be defined as the ability to make use of simulated mental imagery to solve problems—perceiving, discriminating, manipulating, and recalling nonlinguistic images in the “mind’s eye.” Humans do more than “act” in space; they “cognize” about space (Tommasi & Laeng, 2012). Once the eyes have transmitted visual information, the visual system of the brain automatically performs several low-level computations (e.g., edge detection, light–dark perception, color differentiation, motion detection). The results of these low-level computations are used by various higher-order processors to infer more complex aspects of the visual image (e.g., object recognition, constructing models of spatial configuration, motion prediction). Traditionally, tests measuring Gv are designed to measure individual differences in these higher-order processes as they work in tandem to perceive relevant information (e.g., a truck is approaching!) and solve problems of a visual-–spatial nature (e.g., arranging suitcases in a car trunk).
Among the CHC domains, Gv has been one of the most studied (Carroll, 1993). Yet it has long been considered a second-class citizen in psychometric models of intelligence, due in large part to its relatively weak or inconsistent prediction of important outcomes in comparison to powerhouse abilities like Gf and Gc (Lohman, 1996). But “the times they are a-changing.” Carroll (1993), citing Eliot and Smith (1983), summarized three phases of research on spatial abilities, ending in large part in the late 1970s to early 1980s (Lohman, 1979). A reading of Carroll’s survey conveys the impression that his synthesis reflects nothing more than what was largely known already in the 1980s. We believe that the Gv domain is entering a fourth period and undergoing a new renaissance, which will result in its increased status in CHC theory and eventually in cognitive assessment. Carroll, the oracle, provided a few hints in his 1993 Gv chapter.
Carroll (1993) was prophetic regarding two of the targets of the resurgent interest in Gv and Gv-related constellations (often broadly referred to as spatial thinking, spatial cognition, spatial intelligence, or spatial expertise; Hegarty, 2010; National Research Council, 2006). In Carroll’s discussion of “other possible visual perception factors” (which he did not accord formal status in his model), he mentioned “ecological” abilities (e.g., abilities reflecting a person’s ability to orient the self in real-world space and maintain a sense of direction) and dynamic (vs. static) spatial reasoning factors (e.g., predicting where a moving object is moving and when it will arrive at a predicted location).
Carroll’s ecological abilities are reflected in a growing body of research regarding large-scale spatial navigation. Large-scale spatial navigation is concerned with finding one’s way, or the ability to represent and maintain a sense of direction and location, and move through the environment (Allen, 2003; Hegarty, 2010; Newcombe, Uttal, & Sauter, 2013; Wolbers & Hegarty, 2010; Yilmaz, 2009). Using a map or smartphone GPS system to find one’s way to a restaurant, and then to return to one’s hotel room, in an unfamiliar large city requires large-scale spatial navigation. A primary distinction between small- and large-scale spatial abilities is the use of different perspectives or frames of reference. Small-scale spatial ability, as represented by traditional psychometric tests on available cognitive or neuropsychological batteries, involves allocentric or object-based transformation.
Large-scale spatial ability typically involves an egocentric spatial transformation, in which the viewer’s internal perspective or frame of reference changes regarding the environment, while the person’s relationship with the objects do not change (Hegarty & Waller, 2004; Newcombe et al., 2013; Wang, Cohen, & Carr, 2014). Recent meta-analyses indicate that large-scale spatial abilities are clearly distinct from small-scale spatial abilities, with an overall correlation of approximately .27. In practical terms, this means that the ability to easily solve the 3D Rubik’s cube may not predict the probability of getting lost in a large, unfamiliar city. Also supporting a clear distinction between the two types of spatial abilities is developmental evidence suggesting that large-scale spatial abilities show a much faster rate of age-related decline, and that the two types are most likely related to different brain networks (Newcombe et al., 2013; Wang et al., 2014).
The distinction between static and dynamic spatial abilities is typically traced to work by Pellegrino and colleagues (Hunt, Pellegrino, Frick, Farr, & Alderton, 1988; Pellegrino, Hunt, Abate, & Farr, 1987) and is now considered one of the two primary organizational facets of spatial thinking (Uttal, Meadow, et al., 2013). Static spatial abilities are well represented by standard tests of Gv (e.g., block design tests). Dynamic and static spatial tasks differ primarily by the presence or absence of movement. “Dynamic spatial ability is one's ability to estimate when a moving object will reach a destination, or one's skill in making time-to-contact (TTC) judgments” (Kyllonen & Chaiken, 2003, p. 233). The ability to catch a football, play a video game, or perform as an air traffic controller requires dynamic spatial abilities, as “one must note the position of the moving object, judge the velocity of the object, anticipate when the object will reach another point (e.g., one's hand, car, or ship), and take some motor action in response to that judgment. In the perception literature, the research surrounding this everyday human information-processing activity has been known as ‘time to collision’” (Kyllonen & Chaiken, 2003, p. 233). Although the dynamic–static distinction has gained considerable traction and support (Allen, 2003; Buckley, Seery, & Canty, 2017; Contreras, Colom, Hernandez, & Santacreu, 2003), some research has questioned whether the underlying difference reflects an actual spatial ability distinction. [AU: Any update on status of Buckley et al.? No] Kyllonen and Chaiken (2003) reported research suggesting that the underlying cognitive process involved in performing dynamic spatial tasks may be a nonspatial, counting-like clock mechanism—temporal processing, not spatial.
The driving forces behind the increased interest and new conceptual developments regarding spatial thinking are threefold. First, rapid technological changes in the past decade have now made access to relatively cheap and accessible visual-graphic-based technology available to large portions of the population. Individuals can immerse themselves in 3D virtual-reality environments for pleasure or learning. Computer visualizations, often available on smartphones and computer tablets, can be used to teach medical students human anatomy and surgery. The complexities and nuances underling “bid data” can now be unearthed with complex visual network models than can be rotated at will. Anyone can learn geography by zooming over the world via Google Earth to explore locations and cities. Individuals rely on car- or phone-based GPS visual navigation systems to move from point A to point B. Clearly, developing Gv abilities (or spatial thinking) is becoming simultaneously easier via technology, but also more demanding as humans must learn how to use and understand Gv graphic interface tools that present complex visual displays of multidimensional information.
Second, ever-increasing calls have been made to embed spatial thinking throughout the educational curriculum—“spatializing” the curriculum (Newcombe, 2013)—to raise the collective spatial intelligence of our children and youth (Hegarty, 2010; National Research Council, 2006). The extant research has demonstrated a significant link between spatial abilities and educational performance in the fields of science, technology, engineering, and mathematics (STEM; Buckley et al., 2017; Hegarty, 2010; Lubinski, 2010; Newcombe et al., 2013). Gv abilities and individuals with spatially oriented cognitive “tilts” (Lubinksi, 2010) are becoming increasingly valued by technologically advanced societies. More important, research has demonstrated that spatial abilities or strategies are malleable (National Research Council, 2006; Tzuriel & Egozi, 2010; Uttal, Meadow, et al., 2013; Uttal, Miller, & Newcombe, 2013).
Although many psychologists are important drivers of the renewed interest in an expanded notion of the conceptualization and measurement of Gv (e.g., Allen, 2003; Hegarty, 2010; Kyllonen & Chaiken, 2009; Kyllonen & Gluck, 2003; Lubinski, 2010; Uttal, Miller, et al., 2013; Wang et al., 2014), some of the more active research and conceptualizing are being driven by researchers in education (e.g., National Research Council, 2006; Yilmaz, 2009), cognitive neuroscience (e.g., Thompson, Slotnick, Burrage, & Kosslyn, 2009; Wolbers & Hegarty, 2010), and the STEM disciplines (Harle & Towns, 2010; Seery, Buckley, & Delahunty, 2015). Clearly the CHC model’s “mind’s eye” (Gv) is achieving more prominence, which needs to be supported with renewed research on yet to be identified well-supported additional narrow abilities and innovative measurement methods, particularly regarding large-scale and dynamic spatial abilities.
Do other Gv narrow abilities exist? Of course. As with all CHC domains, the validated narrow abilities in the current taxonomy are largely the result of bottom-up programs of research predicated on developing tests for practical purposes (e.g., prediction, diagnosis). Recent conceptualizations of Gv as a broader spatial thinking construct; the dynamic versus spatial and large-scale versus small-scale conceptualizations; and other functional family conceptualizations of Gv abilities are opening a potential Pandora’s box of hypothesized new Gv narrow abilities. For example, Buckley and colleagues (2017) have proposed a comprehensive Gv taxonomy that includes the current Gv abilities and posits 16 potential new narrow abilities based on either theory or research, some previously reviewed by Carroll (1993). These possible new narrow abilities are related to classic spatial tasks (spatial orientation); imagery (quality and speed); illusions (shape and direction, size contrast, overestimation and underestimation, frame of reference); judgments (direction, speed, movement); and dynamic versions of current Gv abilities (visual memory, serial perceptual integration, spatial scanning, perceptual alternations).
These new Gv conceptualizations are welcomed, but they must be studied with serious caution. All new candidates for Gv abilities will need to be validated with well-conceptualized structural validity research (see “Criteria for Updating CHC Theory,” above). Also, if new Gv abilities are identified, it is important to determine whether they have any practical use or validity. An instructive example is a recent CFA CHC-designed study that provided preliminary support for a narrow ability of face recognition (called face identification recognition by the researchers), distinct from other Gv and CHC abilities (Gignac, Shankaralingam, Walker, & Kilpatrick, 2016). The face recognition ability may have practical usefulness, as it could facilitate measurement and research regarding the phenomenon of prosopagnosia (in which a cognitively capable individual is completely unable to recognize familiar faces). Although it is important to guard against premature hardening of the CHC categories (McGrew, 2005; Schneider & McGrew, 2012), we believe that even greater due diligence is necessary to prevent premature proliferation of new entries in the Gv domain in the CHC model. We don’t want to be at a place soon where formal START negotiations (STrategic Ability Reduction Talks) are necessary to halt unsupported speculation about and proliferation of Gv abilities.
- Posted using BlogPress from my iPad