Averaging synhesized voices from multiple speakers increases the attractiveness and smoothness of the averaged voice. In the experiment, we asked 20 participants to rate the attractiveness of 80 voices. The results show that the more speakers are used to average the voice, the more attractive and smoother the voice is. Here, you can listen to examples of voices averaged from differet amount of synthetic voices.

The covariance effect The larger the covariance between two numeric choice options, the more discriminable the choice options are and the higher the probability that the option with the larger expected value is chosen. The figure below presents two numeric options that are dependent on two events. The probability of occurence of each of the events is 50%. In case of Event 1, the payout of Option A is 10 points and of Option B 11 points. By clicking on the scale, you can manipulate values of the standardized covariance. You will see, how outcomes of Option B change, depending on the value of the standardized covariance. When the standardized covariance is 0, Option B is a sure thing, whereas when the standardized covariance is large, outcomes of Options A become more similar to the outcomes of Option B. When the standardized covariance equals 1, Options A and B are identical and the difference between their expected values is 0.

Standardised covariance is a measure of the strength of the association, similarity and co-riskiness between choice options. Co-riskiness is high when the risk level (i.e. variance) of two choice options is similar and the co-riskiness is low, when the risk level of two choice options differs substantially. The figure below demonstrates how similarity between outcomes of Options A and B, association and co-riskines between Options A and B change depending on the value of the standardized covariance. How to use the demo Click on different places along the horizontal line. The horizonal line is a scale of the standardized covariance. Depending on where you click, the outcome values of Option B, and the value of the standardised covariance will change.

Decision Field Theory's stochastic decision process The movie presents the cognitive decision process assumed by the Decision Field Fheory (Busemeyer & Towsend, 1993). Imagine a situation, in which a decision maker chooses between two risky options A and B, for example two car insurance offers. Each option has four attributes. In the car insurance case, the attributes could be: Attribute #1 - price, Attribute #2 - coverage in case of an accident, Attribute #3 - inclusion of assistance programme, Attribute #4 - costumer service. Over time, the decision maker stochastically shifts their attention from one attribute to another and samples information about Options A and B. Some information support choice of Option A, some of Option B. The decision is taken when the accumulated information passes the decision threshold, defined by theta. If there is insufficient information to pass the threshold, the decision is taken when the deliberation time is over. In this case, the decision maker chooses the option in favour of which more information is accumulated. In the movie, attributes to which a decision maker pays attention light in red. There are two decision thresholds, each for one option. This simulation shows three stochastic processes of information sampling.