Tutor Definition: A total of 12 lessons developed for the SimBioSys genetics tutor project at CMU, developed primarily by Albert Corbett (HCII), Benjamin MacLaren (HCII) and Linda Kauffman (Dept. of Biological Sciences). The plan is to implement 18 lessons so that approximately half of the course taught at CMU is covered at some level.

genetics cognitive tutor screenshot

Interface example from the genetics tutor

The 12 Cognitive Tutor Lessons developed so far:

  1. Mendelian Transmission 1: Forward Modeling: Students generate all possible genotypes (one gene) when two parents of same and different phenotypes are crossed and describe phenotypes and genotypes of resulting offspring. They then summarize possible inferences from offspring phenotypes to parental genotypes for purposes of experimental design and interpretation.
  2. Mendelian Transmission 2: Experiment Design and Interpretation: Students are given three strains with phenotype(s) determined by a single gene. Students design and interpret crosses to determine the genotypes of the three strains.
  3. Pedigree Analysis: Students infer dominance and linkage of rare diseases in pedigrees.
  4. Pedigree Analysis and Offspring Probability: Students infer dominance and linkage of rare disease in pedigree, calculate probability that unaffected individuals are carriers and calculate the probability that an unobserved individual is affected.
  5. Pedigree Probability Explanations: Same as Pedigree Analysis and Offspring Probability and students also explain reasoning behind each probability calculation.
  6. Gene Interaction: Mendelian principles are extended to multiple alleles and to two genes with gene interactions. Students cross three true-breeding strains with different phenotypes, decide how many genes are segregating and identify the genotypes of the various parental strains and offspring strains.
  7. Progeny Prediction: Recombination is introduced. Students are given parental genotypes for two genes and the distance between the genes. Students model recombination and segregation and determine all possible offspring phenotypes and their respective probabilities.
  8. Three-Factor Cross: Students are given the results of a test cross and determine the order of three genes and the map distance between each pair.
  9. Time of Entry: Students are given time of entry graphs for three E. Coli Hfr strains. Students determine the time of entry for each of the markers transferred by the Hfrs and map the markers and origin of transfer of the three Hfrs.
  10. Tetrad Analysis: Students are given the set of unordered tetrads that result from a cross. Students classify the types, reason quantitatively about the types to draw qualitative linkage conclusions and calculate the map distance between the genes.
  11. Gene Regulation 1: Forward Modeling: In a typical problem, a gene regulation system is described, with one or more regulatory proteins and non-protein effectors. Students summarize how the system works in a sentence and then reason about the effects of mutations on gene action.
  12. Gene Regulation 2: Experiment Design and Interpretation – Students are given a regulation system for an unknown protein, and they must determine how it is regulated from the behavior of crosses involving mutations in the system’s genes.

History: In the Fall semester of 2002 the first lesson, on pedigree analysis, was developed as part of the Cognitive Tutor course at CMU.

How the interface was built and how the CTAT tools were used to build the tutor: The lessons were almost all built using NetBeans and Java-based DorminWidgets? described on the Java Interface Widgets page. Because the developers have had fairly extensive experience with developing rule-based systems and tutors already, the tool was primarily used for debugging rules (the rule-instantiation window and the behavior recorder).