Desktop version

Home arrow Philosophy

  • Increase font
  • Decrease font

<<   CONTENTS   >>

Future directions

Now that we have addressed the historical and current state of the field, we turn our attention to the future. Although research on levels of processing has spanned nearly a half-century since the work of Marton and Saljo, some progress has been made to create more sophisticated models and frameworks of levels of processing. However, we believe that we have much further to go. We now offer our suggestions for future research and implications for practice.

Future Directions for Research

First, as is evident from these reviews, there has been very little cross-theoretical work in the area of levels of processing. SAL has remained primarily a European and Asian framework, while SRL and the MDL have been primarily used in North America. The SAL tradition, which focuses on the role of the environment on levels of processing, has failed to meaningfully incorporate individual difference factors as well as specific task-level variables in its research agenda. Conversely, both SRL frameworks and the MDL have not taken the role of the learning environment appropriately into account. Some fusion of these primarily endogenous and primarily exogenous approaches to researching levels of processing is needed. As this research continues, it will be vital to continue to refine and adjust our definitions of surface-level and deep-level processing to address the challenges of modeling both individual and contextual factors of depth of processing on performance.

Second, it is also evident that new measures and measurement techniques will be required to propel the field forward. Fortunately, while there is still a long way to go, efforts are already underway to work collaboratively to solve these issues. One such effort is the scientific research network, Learning Strategies in Social and Informal Learning Contexts, which has a major focus on measures and measurements that expand our repertoire of tools including eye tracking (Catrysse et al., 2018), heart rate (Sobocinski, Malmberg, Jarvela, & Jarvenoja, 2018), and neurobiological tools such as functional near-infrared spectroscopy (Dinsmore, Fox, Parkinson, & Bilgili, 2019). Although these are exciting approaches, it remains to be seen how the plethora of data generated can be effectively analyzed - or, as we will discuss subsequently, how these data may be useful to practitioners. The reader is directed to latter chapters of this Handbook for suggestions on how this might occur (Cho, Woodward, & Afflerbach, this volume; Freed, Greene, & Plumley, this volume; Fryer & Shum, this volume).

Implications for Practice

Past research - and these reviews in particular - offer less guidance on future implications for practice. A notable exception to this trend is offered by * Afflerbach’s (2008) review. While not specifically geared toward levels of processing, the review article that was written for reading practitioners (e.g., teachers) would offer a blueprint for discussing the role of levels of strategies for teachers across disciplines. For instance, providing a detailed conception of how different strategies within science (Lombardi & Bailey, this volume) could be considered at the levels of strategies discussed here would provide a service to the field. There are few materials available for practitioners that specifically discuss the issues of levels of strategy use, with a few exceptions (e.g., ’Dinsmore et al., 2018).

The bigger issue, however, is providing teachers with tools to measure students’ levels of processing in any systematic way or on a more mass scale. The time and labor-intensive processes to collect, transcribe, and code think-aloud protocols are not realistic for teachers; neither are the data-intensive processes to analyze the myriad of strategies used by students on a daily basis approachable for teachers. Again, although we have far to go in this regard, there are potential solutions available in related areas of research.

In two ways, technology can be a helpful asset here. First, technology can be helpful in collecting these data. A good example of this trend is Fryers application to measure interest (Fryer & Nakao, 2018). The application uses QR codes to scan and record interest levels in participants for certain tasks and activities. This idea has the potential to be exploited for use to measure levels of processing as well. Rather than rely on verbal transcriptions, students could be trained to use an application to concurrently report their strategy use and the levels of that use. Second, some system would be needed to help teachers analyze that data. In many areas of research a promising avenue to solving this problem is with machine learning (Pereira, Mitchell, & Botvinick, 2009). Machine learning uses the powerful computer processing and artificial intelligence to try to analyze patterns in data that humans cannot. However, we certainly have a long way to go with regard to designing and testing systems to analyze students’ levels of processing.

Concluding Thoughts

We believe this is an exciting time to be engaged on research dealing with levels of strategy use. The reviews contained in this chapter point to many promising avenues of research that can enable us to better understand how these levels of processing are influenced by individual and contextual differences, how those differences might mediate and moderate the relation between levels of processing and performance, and finally, how those levels of processing might directly relate to performance. We hope this review is helpful to those already engaged in the field, but particularly to those new to this area of research.

<<   CONTENTS   >>

Related topics