Analyses of Visual Gaze in Teaching

A central feature of instructional effectiveness is teachers’ responsiveness to the progress of students moment-to-moment. Perceptual acuity plays a key role in teachers’ ability to isolate important student cues, disregard unnecessary information, and respond with instruction that leads to students’ achievement of proximal goals.

It has long been understood that expert music teachers differ from novice teachers on a number of dimensions of behavior, including type and frequency of teacher feedback, pacing, instructional time devoted to student performance, and sequencing (e.g., Cavitt, 2003; Siebenaler, 1997). Highly effective music teachers exhibit similar patterns of instructional behavior that lead to observable improvement in student performance (Duke & Simmons, 2006).

The cognitive processes that guide perception and action form the basis of teacher expertise. Each instructional decision necessarily begins with a teacher’s assessments of the multiple aspects of student performance. The auditory and visual information that teachers attend to (or ignore) sets the stage for all subsequent events of the lesson.

It is often said—and almost as often misattributed to Shakespeare—that the eyes are the windows to the soul. In fact, it may be the case that the eyes are the windows to the mind; that they reliably reflect the focus of human attention and cognition.

During everyday scene perception, the eyes gather precise visual information from only a limited centralized focal point on the retina called the fovea. Given the small region of the visual field that is actually in-focus at any moment, the eyes engage in rapid linear movements called saccades that reorient to multiple fixation points in the field of view, thus allowing the visual cortex to construct an image of the environment that facilitates navigation and the accomplishment of various tasks that enlist nonconscious or volitional control.

Saccades indicate shifts in cognitive attention from one fixation target to another, and fixations (during which the eye remains locked on a visual target) allow the brain to gather relevant visual information that contributes to understanding and guides motor control. Yarbus, in a landmark study, was the first to observe a link between fixation patterns and cognitive goals (Yarbus, 1967). Since that time, researchers have continued to employ advancing technology to understand the relationships between visual gaze, attention, cognition, and behavior (e.g. Corbetta et al., 1998; Henderson, 2003; Eileen Kowler, Anderson, Dosher, & Blaser, 1995).

Newer lightweight, accurate, video-based eye-tracking technology permits reliable recording of human visual gaze patterns in almost any environment. Typical eye-tracking hardware includes stationary (mounted) or mobile headgear that tracks points of gaze and saccadic movements of humans and other animals. Typical headgear uses two video cameras. One camera records the corneal reflection of the subject’s eye at close proximity. The other camera records the scene from the perspective of the subject. Accompanying software then combines the corneal and scene recordings to produce one video that superimposes a visual representation of the subject’s gaze on the scene, showing precisely where the subject looks. Video analysis software isolates and measures fixations and saccades throughout.

The measurement of fixations and saccadic eye movements does not fully illuminate the variety of cortical and subcortical computations that occur at a given moment, but the data provide information that permits reasonable inferences about human perception and cognition in myriad environments. It is now understood, for example, that saccades between fixations are preceded by attention shifts to the target location (Gottlieb, 1998; Kowler et al., 1995) and that changes in attention are reflected in fixation changes (Corbetta et al., 1998). Saccadic eye movements often reflect planning, spatial memory, reward, and sensitivity to target probability (Kowler, 1990; Platt & Glimcher, 1999; Stuphorn, Taylor, & Schall, 2000).

A number of authors who have studied visual gaze in everyday tasks like making a sandwich or a cup of tea (Hayhoe, Shrivastava, Mruczek, & Pelz, 2003; Land, Mennie, & Rusted, 1999) have found that participants almost exclusively fixate on task-relevant objects while pursuing a goal. This and similar research has identified “visual routines” that reflect a sequential understanding and mental models related to various tasks (Hayhoe, 2000; Roelfsema, Lamme, & Spekreijse, 2000). Hayhoe argues, “people must learn where to look in a scene to get the information they need for component sub-tasks. They must learn the structure and dynamic properties of the world in order to fixate on critical regions at the right time” (Hayhoe, Droll, & Mennie, 2007).

Teaching is an extremely complex task that involves multiple variables, some of which are predictable, others less so. Effective teaching relies on the teacher’s having a clear mental model of the desired outcome, knowing what is important to look and listen for, and the ability to anticipate student behavior.

In our lab, we examine the perceptual processes that occur during instrumental music instruction. We employ eye-tracking technology to characterize the ways in which teachers’ visual gaze is reflective of proximal performance goals.

CML LAB MEMBERS

Doctoral Students