Cooperative Group Learning in Undergraduate Neuroscience: Using Simulations to Complement Problem-Solving Assignments

Abstract

The incorporation of active learning improves student learning and persistence compared to traditional lecture-based teaching. However, there are numerous active learning strategies and the degree to which each one enhances learning relative to other techniques is largely unknown. I analyzed the effectiveness of the addition of simulations to cooperative group problem-solving assignments in an undergraduate 400-level neurobiology course. One section of the course carried out group problem-solving alone, whereas the other section used neuroscience simulations (Neuronify) as part of the problem-solving assignments. Overall, both groups of students learned course concepts effectively and did not differ in their performance on exams or specific exam questions related to the assignments. Students perceived that the assignments and simulations were helpful in their understanding of course material but did not overwhelmingly recommend including simulations in the future. Students using simulations were more likely to report gaining experience with experimental design, and this may be an effective way to build scientific reasoning in non-laboratory courses. However, student frustration with technology was the primary reason that students reported dissatisfaction with the simulations. Overall, cooperative group problem-solving with or without simulations is very effective at helping students learn neuroscience concepts.

Keywords: Neuronify; cooperative group learning; problem solving; simulations.

Figure A1

Neuronify screen showing how neuron properties can be altered. In addition to changing the threshold and resting membrane potential, the duration of the refractory period and capacitance and resistance of the membrane can also be altered.

Figure A2

The set-up of the summation simulation in Neuronify. Activation of touch receptors in the bottom half of the simulation can be used to illustrate both spatial and temporal summation. Neuron characteristics can be modified, and neurons (excitatory or inhibitory) can be added or eliminated.

Figure A3

The set-up of the lateral inhibition example in Neuronify. Blue neurons are excitatory and pink neurons are inhibitory. Neuron properties can be modified, and axonal connections can be added or eliminated.

Figure A4

Example of direct basal ganglia loop model built in Neuronify. The activity of motor cortex neurons is measured; their action potential frequency will drastically increase if inputs to the striatum are activated by clicking on the touch receptors (orange squares).

Figure 1

Student performance on all four exams in the course, including the cumulative final exam. Although the average was slightly higher for the PROB section on all four exams, there were no significant differences between sections on any exam. Error bars represent one standard error of the mean.

Figure 2

Student assessment of the degree to which problem sets improved their understanding of course material, ability to apply concepts, ability to talk with others, and test performance, as well as their perception of being worthwhile, frustrating, and recommended for the future. Ratings were on a 5-point Likert scale (0 = strongly disagree, 2 = neutral, 4 = strongly agree). PROB students reported that the problem sets increased their ability to talk with others about course concepts more than SIM students did. Error bars represent one standard error of the mean.

Figure 3

SIM student assessment of the degree to which simulations in particular improved their understanding of course material, ability to apply concepts, and test performance, as well as their perception of being worthwhile, frustrating, and recommended for the future. Ratings were on a 5-point Likert scale (0 = strongly disagree, 2 = neutral, 4 = strongly agree). Error bars represent one standard error of the mean.