Relational Reasoning in Mechanical Engineering

• In Event: 52.031 - Relational Reasoning in STEM Domains: What Empirical Research Can Contribute to the National Dialogue

Abstract

Purpose/Theoretical Framework

Relational reasoning, or the ability to discern meaningful patterns within a stream of information (Alexander & the DRLRL, 2012) is fundamental to cognitive functioning (e.g., Holyoak, 2012). Manifestations of relational reasoning have been empirically linked to achievement in a variety of STEM domains such as mathematics (Richland & McDonough, 2010), chemistry (Bellochie & Ritchie, 2011; Trey & Khan, 2008), meteorology (Trickett, Trafton, & Schunn, 2009), and medicine (Dumas, Alexander, Baker, Jablansky, & Dunbar, 2014). However, the degree to which relational reasoning ability supports success in the field of mechanical engineering (ME) remains underexplored.
Concurrently, an instructional intervention that trains students to draw relations to previous situations to solve novel problems, called the TRIZ method (i.e., a Russian acronym for “Teoriya Resheniya Izobretatelskikh Zadatch,” meaning theory of inventive problem solving), has been shown to improve solution of challenging engineering problems (e.g., Okudan, Ogot, Shirwaiker, 2006; Orloff, 2006; Vargas Hernandez, Schmidt, & Okudan, 2012). However, while TRIZ instruction is described as a relationally-driven tool (Vargas Hernandez et al., 2012), the association between students’ ability to benefit from TRIZ instruction and their ability to discern meaningful patterns within given information has not been systematically investigated. Therefore, in this study, we addressed that gap. Specifically, we compared students’ performance on the Test of Relational Reasoning (TORR; Alexander et al., 2014) with the novelty and variety of their solutions to complex ME problems as assessed by the TRIZ scoring system, before and after intervention.

Method

Participants for this study were 46 graduate level mechanical engineering students at a large mid-Atlantic university. The Test of Relational Reasoning (TORR, =.82), assessed students’ relational reasoning ability. Novelty and variety of ME problem solutions was determined through an algorithmic formula that designates problem solutions based on their function, physical principle, and working principle (Vargas Hernandez et al., 2012). The variety and novelty of ME problem solutions was assessed both before and after the TRIZ intervention. Participants’ visuospatial working memory capacity was assessed with the complex-span Shapebuilder task (Sprenger et al., in press).


Results

Students produced significantly more varied and novel problem solutions after employing the TRIZ method [F(1, 45) = 8.79, p=0.008, p2 = .33]. Further, score on the TORR was a significant predictor of both the novelty and variety of ME problem solutions after the TRIZ intervention, as well as the degree of change in the novelty and variety of solutions over the course of the TRIZ intervention [β=0.10, t=2.86, p<0.014], even when working memory capacity and grade point average (GPA) were controlled.

Significance

The ability to generate novel solutions to mechanical design issues is crucial for students aspiring to enter the mechanical engineering profession (Vargas Hernandez et al., 2012). Our results indicate that this important skill may be undergirded by relational reasoning ability. This finding is significant for researchers working in the field of engineering education, and those interested in cognitive abilities more generally, because it at once identifies a foundational ability for mechanical engineering students, and a further instance of relational reasoning at work.

Authors

• Denis Dumas, University of Maryland - College Park

• Linda Schmidt