Quasi-child logics

Conclusions

Obviously, any attempt to infer relationships among the various logics on the basis of the present data is somewhat risky. It does not seem unreasonable, however, to appeal to extremes for a sketch of boundary conditions. If one can assume a reasonable extrapolation from the present sample to the adult population in general, it appears that:

• (1) ‘Math Logic’ is consistently employed across the various inference patterns by a very small proportion (usually less than 5%) of the population.

• (2) ‘Child's Logic’ — where modus ponens, contrapositive, inverse, and converse are viewed as valid inference patterns — is consistently employed by a large (i.e., about 33%) proportion of the population.

• (3) The various Quasi-Child Logics are used by a non-negligible proportion of the population.

At the youngest age level represented in this sample, the nine-year-old fourth graders, the various ‘Quasi-Child Logics’ described above were used by about 30% of subjects with many subjects (about 20%) consistently using QCL 5. At the other extreme, in the medical school sample, the various Quasi-Child Logics are used consistently by 20% (or more) of subjects with QCL 1 the most commonly employed.