Theories of Causal Learning Essay Example for Free

Theories of Causal Learning Essay In the article entitled â€Å"When Learning Order Affects Sensitivity to Base Rates: Challenges for theories of Causal Learning† written by Ulf-Dietrich Reips and Michael R. Waldmann, theories causal learning was put to test, more specifically the causal-model theory and the associative theories. With this, the causal-model theory â€Å"assumes that learners form a representation of causal models regardless of the order in which learning information is presented† (Reips and Waldmann, 2006). As with the associative theories, the researchers decided to take on the Rescorla-Wagner theory of associative learning. Associative learning theories tend to place cues and outcomes simply for the predictive and diagnostic learning (Reips and Waldmann, 2006). Predictive learning entailed the prediction of symptoms based on the disease presented to them while diagnostic learning required the prediction of diseases based on the symptoms given to them. Thus, it can be said that predictive learning is the forecasting of the effects based on the causes. On the other hand, diagnostic learning is the determination of the causes based on the effects. These two causal learning methods were placed against the variable of base rates. The researchers did the study because in general, they wanted to find out whether the use of base rates was affected by the learning order, whether it is predictive learning or through diagnostic learning. Moreover, they wanted to test whether the base rates were encoded and if this was put to use. Also, they wanted to find out whether the increasing complexity of the ideas would affect the use of base rates. The researchers conducted studies on students coming from the University of Tà ¼bingen. They were given a list of fictitious diseases and their corresponding symptoms with other diseases given more likelihood than the other/s that were in the form of an inverted M. Three experiments were done which involved almost the same procedures with a little alteration for each experiment. The first experiment was done with the purpose of â€Å"finding out whether learners have the competency or correctly incorporating base-rate information regardless of the sequence of learning† (Reips and Waldmann, 2006). The purpose of the second experiment was to Results were later on computed using the ANOVA. The number of participants varied for the three experiments but is similar in terms of the school where they came from, which is the previously mentioned University of Tà ¼bingen. For the first experiment, the participants composed of 24 students who were given participation credit or were paid DM 5, where â€Å"half of this group was randomly assigned to either of the two learning conditions, predictive or diagnostic learning† (Reips and Waldmann, 2006). In the second experiment, there were 32 participants composing mostly of students from the same university who were recruited at the cafeteria. Just the same, each of them was given participation credit or was paid DM 8 for their help with the study. The participants were randomly assigned to either of the two learning conditions. With the third experiment, the participants remained at the number 32 and were randomly assigned to either of the two learning conditions and were also paid DM 10 or were given participation credits for their contribution with the study. The procedure varied from one experiment to another. In the first experiment, the participants were given typed instructions written in the German language. After reading such, they were asked to summarize the instructions and recite them. With an error in the instruction, they were asked to repeat it all over again until they were able to get the instructions right. They were asked to proceed as a â€Å"guest in a special clinic for viro-neuronal tropical diseases for one day† (Reips and Waldmann, 2006). Those under the predictive learning conditions will have to predict the symptoms based on the disease given and those under the diagnostic learning conditions will have to identify the disease based on the symptoms given. They were given hypothetical patients’ record on an index card. The card contains two symptoms on one side and one disease on the other where the predictive learning group saw the disease first before the symptoms and the diagnostic learning group saw the symptoms first. The participants were later asked to rate the likelihood of the occurrence of the disease. In the second experiment, they used the same inverted-M disease but added diseases and symptoms to make them into a total of six diseases and nine symptoms. For this experiment, the researches used the aid of the computer for their learning trials with the use of software called Micro Experimental Laboratory (MLE). They received practically the same instructions but with additional directions on how to use the computer and two new types of questionnaire. The test was self-paced and the experimenter would simply have to press a button to display the items that the participant will have to answer. For the third experiment, the general procedure will be a written general instruction, computer instruction, learning phase with feedback, instruction for the test phase, test phase of diagnostic judgments without feedback, rating questionnaire, and frequency questionnaire. The results of the first experiment were in consistence with the causal-model theory which states that â€Å"learners try to correctly represent causal knowledge regardless of the sequence of the learning input† (Reips and Waldmann, 2006). For the second experiment, the results revealed that the â€Å"participants acquired the base rates of the diseases in all conditions but used them differently in probability ratings depending on the learning condition† (Reips and Waldmann, 2006). The results affirmed the associative theories but were in contrast with the causal-model theory. The third experiment reaffirmed the results and findings of the first and second experiment. In addition to this, they were able to find out that â€Å"the use of base rates in the implicit measure is clearly less pronounced in the predictive learning condition than in the diagnostic learning condition† (Reips and Waldmann, 2006). Moreover, there was a â€Å"pronounced difference in base-rate use after diagnostic but not after predictive learning in the diagnostic ratings† (Reips and Waldmann, 2006). A future possible expansion that is adhered to by the researchers is the question on â€Å"whether the competency to acquire flexibly accessible knowledge is based on the learning phase or on the retrieval phase† (Reips and Waldmann, 2006). Due to the fact that none of the two learning theories, the causal-model and associative theories, encompass the findings of the experiments, it can be said that there is a need for further research regarding the topic to formulate a theory that would be able to hold true for the findings of this current research (Reips and Waldmann, 2006). This has an impact particularly on the causal-model theory which has been disproved by the second experiment and the third experiment (Reips and Waldmann, 2006). In addition to this, the same model is challenged by the fact that people encode base rates but do not always put them to use (Reips and Waldmann, 2006). Both theories may be applied but they can not totally encompass or they lack certain things that might explain the findings of the last two experiments to be able to fully contain them (Reips and Waldmann, 2006). The findings of the research is of great importance and use to the field of education. More specifically, it is helpful in determining the â€Å"suitable learning and training contexts in education† (Reips and Waldmann, 2006). Moreover, the information gathered from this research will be of great use to the field of medical sciences since the presentation of information with regard to causes may present deficiencies when this is used (Reips and Waldmann, 2006). The findings of this study will have to enhance the over-all learning process and theories that are in current use. References Reips, U. and Waldmann, M. 2006. When Learning Order Affects Sensitivity to Base Rates: Challenges for Theories of Causal Learning. PLEASE PUT THE PLACE OF PUBLICATION HERE: Hogrefe and Huber Publishers.