Desktop version

Home arrow Philosophy

  • Increase font
  • Decrease font

<<   CONTENTS   >>

Discussion and paths for future research

Eye movements, and more specifically second pass fixation durations, reflect partly what students report on their processing strategies. Deeper cognitive processing, as referred to in eye movement research (Ariasi & Mason, 2011; Holmqvist & Andersson, 2017; Penttinen et al., 2013), is not the same as deep processing, as defined by theoretical models on students’ processing strategies (Dinsmore, 2017; Vermunt & Donche, 2017). Deep processing, as defined in theoretical models on processing strategies, refers to the intention to understand what the author wants to say in the text, to engage in meaningful learning, to relate the content of the text to a wider context and prior knowledge, and to focus on the main themes and key information in the text (Dinsmore, 2017; Vermunt 8< Donche, 2017). The study of Catrysse et al. (2016) showed that when students received reproduction-oriented questions that they processed details in the text more thoroughly than students who received questions aimed at deeper processing. This study showed that longer second pass fixation durations can also be related to surface processing strategies and, thus, do not always reflect the deep processing strategies as referred to in theoretical models on processing strategies. Other research provided evidence that longer second pass fixation durations do not solely reflect deep processing strategies. Longer second pass fixation durations are an indication of the multiple use of processing strategies, namely, combining both surface and deep processing strategies (Catrysse et al., 2018), and can be a reflection of deeper processing strategies only in combination with a high topic interest for key sentences (Catrysse et al., 2018). These findings are shown in Figure 19.1 and propose a framework for future research with psychophysiological measures. The most central circle represents the focus of attention. The second dotted circle represents task-specific learner and contextual characteristics, such as students’ interest for a learning task, their processing strategies used during a learning task and the assessment demands for a learning task, among others. The interplay of these characteristics affects the focus of attention as measured with eye tracking. The outer circle represents more general learner and contextual characteristics such as the general disposition towards processing strategies. Both circles interact with each other and affect the focus of attention. Therefore, the inner circles are represented with dotted lines, because all these characteristics interact and may affect the focus of attention. The findings described in this chapter call for two important and related actions for future research: (1) students’ processing strategies is a complex and multidimensional construct and should be measured in that way, and (2) psychophysiological measures need to be applied in multi-method designs. Figure 19.1 can thus be compared with a bull’s eye. In order to measure students’ processing strategies more accurately, the complexity and multidimensionality of processing strategies need to be taken into account for multi-method designs. We further elaborate on these two points below.

Eye Movements and Students' Processing Strategies

Figure 19.1 Eye Movements and Students' Processing Strategies

Based on the research discussed in this chapter, we want to stress the importance of adapting multi-method designs in order to grasp the full complexity of students’ processing strategies and to avoid relying on one single method or instrument. Processing strategies and many learning processes are complex and multidimensional in nature (Alexander, 2017) and, therefore, it is impossible to capture all aspects of processing strategies with one single method. More specifically, research discussed in this chapter demonstrated the added value of applying multi-method designs in order to interpret data from psychophysiological measures. We thus want to stress the importance of combining these measures in order to understand the underlying reasons for processing behavior. By combining different methods, the power of each method is taken to obtain a comprehensive picture and deep insight into students’ processing strategies (Schellings, 2011; Schellings & van Hout-Wolters, 2011). We hereby also want to emphasize that no instrument has a greater value than another; however, it is the combination of methods that results in the greatest strength.

Self-report measures such as verbal reports are mostly used in combination with eye movement data (van Gog & Jarodzka, 2013). In the discussed research, the multiple use and the nature of processing strategies was added as a predictor for explaining differences in eye movements. This research showed that both general self-report questionnaires and cued retrospective think-aloud protocols allowed an explanation of the differences in students’ eye movements and their learning process. Moreover, a recent review of Dinsmore (2017) indicated that how well a strategy is used and how appropriate the chosen strategy is, are better predictors of learning outcomes than the measures that are mostly used. It would be interesting for future research to see how these aspects of students’ processing strategies can explain the differences in eye movements and to verify whether these better predictors for learning outcomes are also good predictors to explain differences in the micro-processes of learning.

In eye movement studies, mostly first pass and second pass duration measures are used to analyze eye movement data (Ariasi et al., 2017; Catrysse et al., 2018; Hydna et al., 2003; Kaakinen & Hydna, 2007; Yeari et al., 2016). However, we are convinced that measures other than duration measures would be interesting to examine as well. As indicated in the work of Holmqvist and Andersson (2017), common analyses for reading texts and textbooks are the distribution of fixation times in areas of interest, sequence orders of areas of interest and transitions between areas of interest. Former self-report research has shown that deep processing not only refers to deeper processing of the key information in a text but also to integrating information across the text (Dinsmore & Alexander, 2016; Fox, 2009; Pressley & Afflerbach, 1995). Therefore, analyzing transition matrices would be a promising avenue to further investigate the relation between processing strategies and eye movements. Transitions are movements between areas of interest and are counted for pairs of areas of interest (Holmqvist & Andersson, 2017), for instance the number of transitions from a key sentence to another sentence. Whether key sentences are used as anchor points to process the rest of the text or not could be analyzed. This kind of analysis could clarify whether typically deep processors use key sentences as anchor points and whether, typically, surface processors use detailed sentences as anchor points in order to build up their mental representation of the text. Therefore, we suggest that future research should look further into this kind of analysis as well and go beyond solely analyzing eye movement duration measures.

For future research it would be interesting to further explore the multidimensional nature of students’ processing strategies. As shown in Figure 19.1, levels of processing interact with task-specific learner and contextual characteristics. Students’ processing strategies are measured in a more dynamic way with psychophysiological measures, but the other learner characteristics, such as for example interest, are up to now mostly measured in a rather static way (Catrysse et al., 2018). Constructs such as students’ interest, however, are in fact clearly dynamic in nature and are expected to fluctuate during the learning process (D’Mello, Lehman, Pekrun, & Graesser, 2014). It is assumed that this dynamic response of the reader has an impact on how attentional resources are allocated during reading (Kaakinen, Ballenghein, Tissier, & Baccino, 2018; Kaakinen & Hydna, 2014). Future research could further explore how taskspecific learner characteristics, measured in a dynamic way with psychophysiological measures, interact with processing strategies while reading and/or learning. By doing so, research can shed light on the multidimensional and dynamic nature of processing strategies and its influencing characteristics.

<<   CONTENTS   >>

Related topics