Page 170 - Contributed Paper Session (CPS) - Volume 4

CPS2165 Carla Susete Gonçalves Francisco et al.
analysis, factor analysis and data adjustment. In order to perform the data
analysis, it will be necessary to use the software R and RStudio (R CoreTeam,
2017) and their respective packages for data visualization and factorial
analysis. The use of spreadsheet software was also required for some data
processing. The research was carried out based on existing theories,
through the analysis of a model that collects all the advantages of different
approaches. This theoretical model was used as a basis for obtaining
simulated data, which was later used to compare the recollected data with
the actual experimental design models. In order to obtain real data, two of
the following sources where used: Preexisting sources or real data. Pre-
existing sources provide us with data of experiments already carried out,
such as the one located in the following website:
https://www.humanbenchmark.com/. Real data was obtained by
conducting some direct experiments, from a website, in order to model and
compare both response times, from the simulated data and the actual data.
All these experiments have to be done by using small samples, since the
results obtained present some problem regarding the size of the datasets
used, however possible in later developments, the datasets can be extended
to obtain more accurate results.

4. Discussion and Conclusion
The results of this study is try to explain the asymmetry of observed
latency distributions. There are several possibilities to make a significant
contribution to these studies:
• Adjust the best distribution analysis for the real data:
• Fieller distribution implies linearity.
• Model LATER-d implies non-linearity.
• Recinormality assessment: The evaluation of the recinormality of
the data sets by item will be provided by the analysis of the
Reciprobit graphs.
• Separation of the effect stop-down of the bottom-up: distinguish
between the variations of the intercept and the slope being the
latter the top-down effect. Such as manipulations of the response
prepriest probability to manipulate the difficulty of perceiving a
stimulus.
• Zone recinormal: This analysis assumes that the data are normally
distributed, that is, they fall into the Recliner zone of the Fieller
distribution (λ1 < 0.22). But some assays can lead to lambda values
between: (0.22 < λ1 < 0.4). Therefore, we could determine the
proximity to the Recinormal zone.

159 | I S I W S C 2 0 1 9