Video feedforward for reading.

Video feedforward can create images of positive futures, as has been shown by researchers using self-modeling methods to teach new skills with carefully planned and edited videos that show the future capability of the individual. As a supplement to tutoring provided by community members, we extended these practices to young children struggling to read. Ten students with special needs participated in a multiple baseline intervention. Each received tutoring only, followed by tutoring plus video feedforward, another phase of tutoring only, and follow-up. Overall, reading fluency improved significantly for all students; in 9 out of 10 cases, rate of improvement was significantly greatest during feedforward. Other measures (e.g., word identification) confirmed student progress from most at-risk to mid-stream status. We conclude that video images of success with challenging materials may enhance the acquisition of reading skills.

**********

Self-modeling has been studied most explicitly in the video medium. Approximately 300 such applications in which the participant is both the model and the observer have been described in print (see review by Dowrick, 1999). Typically, participants watch themselves in brief videos in which they perform successfully in challenging situations (self-modeling terminology defined by Dowrick & Raeburn, 1977, 1995). Many applications have been with school-age children, often in special education and related services (e.g., Bray & Kehle, 1998; Buggey, 1995; Walker & Clement, 1992), but not many applications have addressed academic performance (see Hitchcock, Dowrick, & Prater, 2003). Exceptions include studies in the acquisition of mathematical skills by Schunk and Hanson (1989) and Woltersdorf (1992).

The successes to be observed on video may be selected as exemplars taken out of the context of typical performance, as is often done for classroom on-task behavior. For example, Kehle, Clark, Jenson, and Wampold (1986) recorded hours of unprompted classroom behavior of four elementary school children (boys) who had been identified with behavior disorders, and then they edited a selection of positive examples into relatively short videos. Each student watched his own videotape once a day for 5 days, and the rates of inappropriate behavior declined, on average, from 47% to 11%. This type of self-modeling--to increase adaptive behavior that is currently intermixed with nondesired behaviors--is known as positive self-review (PSR). PSR is also used for mood-based disorders, the transferal of role play to the real world, and the maintenance of disused or low-frequency skills (Dowrick, 1999).

All creatures learn from observing their successes, but not usually on video; humans distinguish themselves by being able to learn through observing successes they have not yet had. This type of self-modeling is known as feedforward, an image of future mastery (a term coined to contrast with feedback, which illustrates past or present performance). For example, behavior that occurs only in one setting may be transferred to other settings by video or audio feedforward (Blum et al., 1998). In the classic application to selective mutism (Dowrick & Hood, 1978; also described in Dowrick, 1999; Kratochwill, 1981), two children were never observed to speak at school, although they spoke freely at home. Essentially, videos were made in both locations, and each child's conversation from the home was edited into the context of an adult asking questions and making comments to the same child at school. Both children watched together--first one feedforward video over a period of days, then the other--with the result that each child significantly increased the rate of speech at school in response to the self-model but not to the peer model. These increases in speech were rapid and dramatic, with only a few viewings of the feedforward video.

Other categories of self-modeling for which a feedforward approach is effective include using visually hidden support in the situation of anxiety-impaired performance, such as learning to swim (e.g., for children with spina bifida; Dowrick & Dove, 1980). Another example is the strategy of combining component skills, as in separately performed elements of figure skating (take off, spin, landing), to produce the video image of a routine, such as the triple lutz, that has yet to be mastered (Dowrick, 1989). Thus, video feedforward has demonstrated efficacy in the acquisition of physical skills, social skills, and classroom behavior. Yet, as noted, few studies have addressed the learning of basic academics most essential in special education.

Literacy is fundamental to all education in most schools. In 2002, the national reading scores for fourth-grade students were about the same as in 1992, but the spread of scores was even greater (NCES, 2003). Disadvantaged urban schools most frequently report the worst literacy outcomes. Schools with which we have worked in the past 7 years, in low-income areas of different states, have experienced needs far beyond their resources. About 90% of the children qualified for free or reduced-priced lunch. Over all sites, 80% of the children were from immigrant, African American, or Native Hawaiian families, and they lived in the lowest 10% of housing amenities in their cities (estimate based on housing types/values and family income); 35% of adults had no high school education (cf. 7.5% nationally; U.S. Census Bureau, 2000). The schools were in the bottom 20% of their districts for achievement in basic academics, and 12% to 25% of students were classified in special education (Hawai'i Department of Education, 1999; University of Pennsylvania Library, 1998).