Desktop version

Home arrow Philosophy

  • Increase font
  • Decrease font


<<   CONTENTS   >>

Generality and specificity within and across individual students

The focus of this chapter is on the enactment of strategic processes both within and across domains of learning, wherein a strategy that is utilized across multiple domains can be described as domain-general, but a strategy that is only utilized within a single domain can be described as domain-specific. However, such a designation begs a follow-up question: are domain-general strategies utilized across domains by the same individual student or are they merely utilized across domains, but by different individual learners? Further, are there individual differences across students in the readiness with which they transfer strategic knowledge to new tasks or domains?

As an example of this general query, take a strategic process that is typically considered to be domain-general, such as connecting to prior knowledge. Theoretically, such a strategy must be considered domain-general because it is easy to imagine that, regardless of the academic situation, new information being presented to a student may be related in some meaningful way to something that the student already knows. Indeed, researchers who have studied students learning across a variety of domains (e.g., Afflerbach, 1990) have observed that connections to prior knowledge can and do arise across domains. However, it is also relevant to consider that, within the same student, certain domains of learning may appear more salient or relevant to their past experiences for a variety of socio-emotional or identity-based reasons (e.g., Hartwell & Kaplan, 2018). Students may be differentially cognitively effective at mapping new information onto their prior knowledge across domains where the relations between prior knowledge and current instruction are not made explicit (Richland & McDonough, 2010), or they may simply possess differential amounts of prior knowledge across domains, limiting the possibility of them connecting new information to that prior knowledge. Therefore, even a highly domain-general strategy such as connecting to prior knowledge can be variant in its generalizability as to its actual usage within a particular student.

This issue is closely related to the question of transfer within the educational and cognitive psychology research area (Marcus, Haden, & Uttal, 2018). In the 2010s, a relatively large quantity of research was published in which researchers attempted to train participants on cognitive functions that are theoretically very domain-general such as working memory (see Melby-Lervag & Hulme, 2013 for a meta-analysis). Of course, the data showed that continued engagement with such cognitive training did substantially improve participants’ performance on the tasks or games on which the participants were practicing (Jaeggi, Buschkuehl, Jonides, & Perrig, 2008). Unfortunately, another resounding finding from this area of research was that the gains in ability that participants displayed were limited to the task on which they practiced, or very similar tasks (Sprenger et al., 2013). So, despite the cognitive training taking place on a task that was designed to measure an entirely domain-general ability, learning gains on that ability did not actually influence domain-learning in the way that was hypothesized. So, does that mean that the abilities trained were indeed domain-general or not?

One possible explanation for this effect that is relevant to the topic of this Handbook is that, in order to improve on cognitive training tasks, participants refined their task-specific strategies. These task-specific strategies may have allowed them to improve their performance on those particular tasks but did not allow for general gains on other tasks that were more nested within typical academic domains. Such a hypothesis highlights an interesting paradox concerning the tasks that are often used by psychologists to measure domain-general cognitive abilities (e.g., visuo-spatial reasoning tasks; Dumas & Alexander, 2016). While these tasks are not nested within a particular academic area and are therefore not highly influenced by prior domain-knowledge, they themselves constitute a sort of domain made up of similar tasks. For this reason, some have suggested that the quantification of general capacities should also be undertaken by examining the higher-order patterns among domain-specific measures, as opposed to only abstract tasks (Dumas & McNeish, 2017).

Within educational research on cognitive strategies, this problem is especially salient because, when we make practical recommendations to teachers, we must contend with the possibility that, although a particular cognitive strategy strongly supported student learning in our data, that strategy may not suffice to improve student performance across the range of tasks that students actually encounter in school and in life. For example, relational reasoning strategies are one body of cognitive procedures that have been empirically connected to student learning outcomes across a wide gamut of academic contexts ranging from elementary reading (Farrington-Flint, Wood, Canobi, & Faulkner, 2004) to medical residency (Dumas, Alexander, Jablansky, Baker, & Dunbar, 2014), and many instances in between. However, it is not yet known whether the fact that we can observe students engaging in relational reasoning across those learning contexts means that relational reasoning instruction, if abstracted from domain-specific academic material, would be effective at improving student performance across many domains (Dumas, Alexander, & Grossnickle, 2013). Although future work is necessary to address this research question, I would hypothesize that domain-general relational reasoning instruction would not necessarily improve student performance across all of the domains in which relational reasoning is known to play a role. Instead, it may be that, over the course of domain-learning, students must develop sophisticated strategies for identifying patterns within the information they interact with (i.e., relational reasoning strategies), and that is why the strategies appear so relevant across domains. In this way, a strategy that appears domain-general may actually have developed within a specific domain for a particular student. This issue is related to a further theoretical area that is relevant to the domain-generality of strategies and skills: the way in which the development of expertise influences learners’ ability to apply strategies across (as opposed to within) domains.

 
<<   CONTENTS   >>

Related topics