Monday, January 30, 2017

Questioning the Dual Channel Theory

I was investigating the dual channel theory used in Mayer and Moreno (1999). After a review of the literature, it appears that the theory is currently unreliable.

Long-term and Working Memory

Cognitive psychology identifies two types of memory: a very limited working memory and a theoretically unlimited long-term memory. Memories stored in long-term memory are stored in “hierarchically organized schemas which permit us to treat multiple sub-elements of information as a single element” (Kalyuga et al., 1999, p. 351). Working memory, on the other hand, is very limited and can be overloaded easily unless attached to an existing schema. Thus, Kalyuga et al. (1999) posit a split-attention effect where some working memory resources must be devoted to processing the connection between two separate pieces of information (a diagram and text explanation, for instance). Kalyuga et al. (1999) indicate the audio presentation of information improves learning by overcoming the split-attention effect since the diagram and audio explanation can be presented simultaneously. Mayer and Moreno (1999) confirm the main premise of Kalyuga et al. through their modality effect. They propose a “dual-processing model of working memory with separate channels for visual and auditory processing” (Mayer & Moreno, 1999, p. 359). Both Kayuga et al. (1999) and Mayer and Moreno (1999) relate their research to previous cognitive psychological theorists who observed some form of splitting in the working memory.

Cognitive Psychology Support for Separate Audio/Visual Channels

Several cognitive psychologists have presented evidence of divisions in working memory. Paivio (1991) has researched the “dual coding theory (DCT) of memory and cognition” (p. 255) for decades. Paivio (1991) posits a split in working memory between verbal, including printed or spoken words, and non-verbal elements, including visual objects or environmental sounds and other sensorimotor impressions. Likewise, Penney (1989) observed a difference in short-term retention of information either seen or heard. Her separate-streams hypothesis, however, only identifies a difference between the modalities that lasts only seconds in a sensory store. As Reinwein (2012) identifies, another theorist, Alan Baddeley, posits a split in working memory between a visuo-spatial sketchpad that processes symbolic information and a phonological loop that processes verbal information. Printed information that is read is also processed through this phonological loop indirectly through an internal articulation. There is evidence of divisions in working memory. However, as Reinwein (2012) argues, while there may be analogous connections between the splits identified in the split-attention effect and the modality effect and the divisions found by Paivio  (1991), Penney (1989), and Baddeley, none of cognitive psychological models presented directly supports separate channels for visual and auditory processing.

Questioning the Research Supporting Separate Audio/Visual Channels

Apart from the lack of support from cognitive psychology for the theories of Kayuga et al. (1999) and Mayer and Moreno (1999), questions have been raised about the reliability and validity of their research. Reinwein (2012) questions several aspects of the research of Kayuga et al. (1999) and Mayer and Moreno (1999). First, he identifies the lack of direct support from the aforementioned cognitive psychologists who Kayuga et al. (1999) and Mayer and Moreno (1999) cite. In fact, Reinwein (2012) indicates that both Kayuga et al. (1999) and Mayer and Moreno (1999) have “experimentally crossed the variables Modality (visual, auditory) with Memory (short term, long term)” (p. 27). Second, Reinwein (2012) performs meta-analysis on the same published research of that another researcher had previously studied in a meta-analysis. Having corrected for certain methodological problems, Reinwein (2012) found half the effect size previously identified by the other meta-analysis. Third, Reinwein (2012) further halves the effect size when correcting for publication bias, where studies showing high effect sizes are more likely to be published than studies showing low effect sizes. Lindow et al. (2011) hypothesize about how the same publication bias may explain how robust the research underlying the modality effect appears, at first. They note that almost half of the published studies supporting the modality effect appear in a single journal, the Journal of Educational Psychology. Lindow et al. (2011) attempt to replicate the findings of Mayer and Moreno (1999) by repeating their experiment and fail to find a modality effect. They hypothesize that, rather than a modality effect, an “auditory recency hypothesis” (Lindow et al., 2011, p. 232) may explain the effect seen in the findings of Mayer and Moreno (1999). The auditory recency hypothesis posits that the final sentence heard is retained significantly longer than when read. Thus, the length of text used in Mayer and Moreno (1999) may have contributed to the appearance of a modality effect. According to Lindow et al. (2011), the questions that have been raised about Mayer and Moreno (1999) and the lack of replication of their work casts doubt on the validity of their work, and more confirmation or disconfirmation in future studies is needed.

Conclusion

Serious questions have been raised about the reliability of the research that indicates a split between an audio and a visual channel in working memory. If these separate channels do not exist, the Modality Principle is brought under question, and if it turns out the Modality Principle does not exist, then the Reverse Modality Effect would need to be renamed.



References

Kalyuga, S., Chandler, P., & Sweller, J. (1999). Managing Split-attention and Redundancy in Multimedia Instruction. Applied Cognitive Psychology, 13, 351-371.

Lindow, S., Fuchs, H. M., Fürstenberg, A., Kleber, J., Schweppe, J., & Rummer, R. (2011). On the robustness of the modality effect: Attempting to replicate a basic finding. Zeitschrift für Pädagogische Psychologie, 25(4), 231-243.

Moreno, R. & Mayer, R.E. (1999). Cognitive Principles of Multimedia Learning: The Role of Modality and Contiguity. Journal of Educational Psychology, 91(2), 358-368. 

Paivio, A. (1991). Dual Coding Theory: Retrospect and Current Status. Canadian Journal of Psychology. 45(3), 255-287.

Penney, C.G. (1989). Modality effects and the structure of short-term verbal memory. Memory & Cognition, 17 (4), 398-422.

Reinwein, J. (2012). Does the Modality Effect Exist? and if So, Which Modality Effect? Journal of Psycholinguistic Research, 41(1), 1-32.