Epistemological beliefs are fundamental assumptions about the nature of knowledge, the certainty of knowing, the criteria and justifications for knowing. There is a variety of views explaining epistemological beliefs such as developmental, multidimensional, and integrated theoretical models. The model of Perry [1], King & Kitchener [2], Baxter Magolda [3] described epistemological developments as a progression, Schommer [4] and Hofer [5] considered multidimensional frameworks of epistemological beliefs. Schommer [4,6] proposed epistemological beliefs consisting of five dimensions: stability of knowledge, structure of knowledge, source of knowledge, speed of knowledge acquisition and control of knowledge acquisition. Hofer [5] indicated four types of beliefs dimensions: knowledge of certainty, source of knowledge, the simplicity and the justification of knowledge. On the other hand, epistemological belief is an individual presupposition about knowledge and learning which play an influential role on the cognitive process [7]. They presented integrated theoretical model between epistemological beliefs and self-regulated learning [7,8].

Previous educational studies have examined the effects of epistemological beliefs on such areas of learning as reading comprehension [6], text processing [9], conceptual change [10] and multiple decision making [11,12]. On the other hand, instructional contexts could change the students’ epistemological beliefs [13,14].

Epstemological beliefs also affect problem-solving skills and strategies in poorly structured environments [6,15,16]. Students who had simpler understandings of the structure of knowledge and invariability of knowledge made simpler diagnoses than did those with more complicated understandings. They also ignored divergent and idiosyncratic patient situations in the decision-making process [17]. Learners with a belief that knowledge is a fixed entity have been shown to engage in more rapid and superficial learning than thoughtful and deep learning [18]. Schommer & Dunnell [19] showed the more students had a belief that the ability of learning is decided at birth, learning is quick, and knowledge is unchangeable, the more likely they had an attitude of writing simplistic answers to problems.

Roex & Degryse [20] described the possible role of epistemological beliefs in medicine as follows: First, the explosive growth in the amount and accessibility of medical information expects the strategies on how to acquire knowledge. Second, ill-structured problem solving often demands solvers to think about various views and apply multiple standards while assessing problems [21,22]. Third, medical practice is a problem-solving process with scientific knowledge. Doctors must understand the origin of the knowledge, integrate new knowledge into existing knowledge and apply the treatment into patient.

Since personal epistemological beliefs affect students’ behaviors and learning achievements, medical students’ clinical performance examination (CPX) achievements may differ by students’ epistemological beliefs stances. We hypothesized that students with naive beliefs such as knowledge is unchangeable and the source of knowledge is from authorities would be less likely to implement effective CPX strategies. Students with sophisticated beliefs about the complexity of knowledge, multiple perspectives on learning, and progressive learning processes would be more likely to apply multifaceted approaches with their patients. Therefore, the purpose of this study was to explore the relationship between medical students’ epistemological beliefs and their achievement on the CPX.

Table 2 shows differences in CPX category scores with respect to overall epistemological beliefs levels. The higher epistemological beliefs group had significantly higher scores for history taking (t=2.285, p=0.024) and patient-physician interaction (t=2.004, p=0.047) than did the lower one. No significant differences between higher and lower epistemological beliefs groups were observed with respect to physical examinations.

Statistically significant correlations were found among the epistemological beliefs scales and CPX category scores (Table 3). The score of epistemological beliefs for certainty of knowledge was significantly positively correlated with those for patient-physician interaction (r=0.250, p=0.005). Epistemological beliefs scores for source of knowledge were significantly positively correlated with those for history taking (r=0.174, p=0.050) and patient-physician interaction (r=0.231, p=0.010). Epistemological beliefs scores for ability to learn were significantly positively correlated with those for history taking (r=0.232, p=0.010), physical examination (r=0.224, p=0.012), and patient-physician interaction (r=0.225, p=0.012).

The higher epistemological beliefs group had higher achievements for history taking and patient-physician interaction than did the lower one. These results could indicate that students with a sophisticated, less dualistic epistemological beliefs stance performed better on the CPX, especially with respect to history taking and the patient-physician interaction.

There were slightly weak-positive correlations between epistemological beliefs scores on certainty of knowledge and CPX scores measuring patient-physician interaction. The weak-positive correlations were also demonstrated between epistemological beliefs scores on source of knowledge and the CPX scores measuring history taking and patient-physician interaction. Ability to learn was positively correlated with history taking, physical examination, and patient-physician interaction.

Although the correlations with epistemological beliefs and CPX components were not very strong, these results were consistent with the findings of Schommer [6] and Cano [23], which showed that students who believed knowledge is not absolute or acquired in an unmediated fashion attained greater academic success. In poorly structured medical situations, students with simpler understandings about the structure of knowledge provided simpler diagnoses than did those who demonstrated more complex thoughts in this regard [17,24,25].

Scores on source of knowledge were related to how students view the patient-physician interaction. Those who had a more developed understanding of the source of knowledge were more likely to question the opinions of authorities when their own understanding differed. These students also paid more sincere attention what their patients were saying, regarding this information as a good source of knowledge. Students should try to identify plausible interpretations of symptoms and signs and could be able to suggest alternative while gathering information from and sharing it with patients.

Additionally, ability to learn of the epistemological beliefs was correlated with physical examination as well as history taking and patient-physician interaction. As physical examination was the item to evaluate the level of technical skills, there might be good results on the behaviors of students who believed the skills and attitude could be enhanced by repetitive practice.

Kienhues et al. [11] investigated how laypersons deal with conflicting versus consistent medical information on the Web. Students with an access to conflicting medical information showed more advanced treatments than those who were provided with well-structured information. Exposure to different ways of thinking could facilitate the beliefs development and may encourage students to become more relativistic and appreciative of multiple perspectives. Our results revealed that more sophisticated and advanced epistemological beliefs were positively related with the clinical performance as the outcomes of learning in medicine. Therefore, educational efforts that encourage discussions pertaining to epistemological views should be considered to improve clinical reasoning and problem-solving competence in the clinical setting.

The limitations of this study should be considered when interpreting the results. First, the correlations between epistemological beliefs and CPX were not strong. Therefore, we should ask more careful interpretations and suggestions on these results. In a further study, various sources would be needed to understand the students’ epistemological beliefs such as interviews and writings. Second, our study focused only on the relationship between medical students’ epistemological beliefs and their CPX performance. Further research may need to be carried out to determine other factors contributing to CPX performance.