Aversive View Memory and Navigational Risk Sensitivity in the Desert Ant, Cataglyphis Velox

Many ant species are able to establish routes between goal locations by learning views of the surrounding visual panorama. Route formation models have, until recently, focused on the use of attractive view memories, which experienced foragers orient towards to return to the nest or known food sites. However, aversive views have recently been uncovered as a key component of route learning. Here, Cataglyphis velox rapidly learned aversive views, when associated with a negative outcome, a period of captivity in brush, triggering an increase in hesitation behavior. These memories were based on the accumulation of experiences over multiple trips with each new experience regulating foragers hesitancy. Foragers were also sensitive to captivity time differences, suggesting they possess some mechanism to quantify duration. Finally, we characterized foragers perception of risky (variable) versus stable aversive outcomes by associating two sites along the homeward route with two distinct schedules, a fixed duration of captivity or a variable captivity duration, with the same mean time over training. Foragers exhibited significantly less hesitation to the risky outcome compared to the fixed, indicating they perceived risky outcomes as less severe. Results align with a logarithmic relationship between captivity duration and hesitation response, suggesting that foragers perception of the aversive stimulus is a logarithm of its actual value. We conclude by characterizing how view memory and risk perception can be executed within the mushroom bodies neural circuitry.

Data Mining Method for Identifying Biased or Misleading Future Outlook

In this study, we introduce a data mining method to identify biased and/or misleading outlooks for future performance of various factors, such as income, corporate profits, production, countries’ GDP, etc. The method consists of several components. One very important component involves building a general model, where the dependent variable is a factor suspected of projecting an over-optimistic impression in some records. Explanatory variables in the model are viewed as representing the potential for the satisfactory performance of the dependent variable. The second component involves evaluating the potential for the individual records of interest (specific countries, corporations, production facilities, etc.), and allows us to identify possible gaps between the upbeat/optimistic projections into the future (of the dependent variable) versus low and/or declining potential. In other words, low and/or declining potential basically tells us that the optimistic future performance of the dependent variable is unattainable, and could also represent misleading or deceitful information. The important novelty of this study is the capability to identify a highly exaggerated outlook of future performance, by utilizing a soft regression tool and the concept of “performance potential”. The process is explained in detail, including the conditions for successful evaluations. Case studies to evaluate expected economic success are presented.

Novel Uses of Three-Parameter Logistic Models and First-Derivative Models for the Coronavirus Disease (COVID-19) Epidemic in the United States, in Three Distinct Scenarios

Background:Forecasting the current coronavirus disease (COVID-19) epidemic in the United States necessitates novel mathematical models for accurate predictions. This paper examines novel uses of three-parameter logistic models and first-derivative models through three distinct scenarios that have not been examined in the literature as of July 14, 2020.Methods:Using publicly available data, statistical software was used to conduct a non-linear least-squares estimate to generate a three-parameter logistic model, with a subsequently generated first-derivative model. In the first scenario a logistic model was used to examine the natural log of COVID-19 cases as the dependent variable (versus day number), on July 11 and May 1. Independent t-test analyses were used to test comparative coefficient differences across models. In the second scenario, a first-derivative model was derived from a base three-parameter logistic model for April 27, examining time to peak mortality and decrease in case fatality rate. In the third scenario, a first-derivative model of mortality through July 11 as the dependent variable, versus confirmed cases, was generated to look at case fatality rate relative to increasing cases.Results:All models generated were statistically significant with R2 > 99%. The logistic models in the first scenario best predicted time to growth deceleration in the natural log of cases in the U.S. (slowing of exponential growth), estimated at March 11, 2020. For the May 1 data, independent t-test analyses of comparative coefficients across models were useful to track improvements from implemented public health measures. The first-derivative model in the second scenario on April 27, when the epidemic was more controlled, showed peak mortality around April 12-13, with a case fatality rate of < 1,000 deaths and trending down. The first-derivative model in the third scenario estimated a near-zero case fatality rate to occur at 4 million confirmed cases. It has not been affected by fluctuations in mortality from June 29 through July 11.Conclusion:Three-parameter logistic models and first-derivative models have utility in predicting time to growth deceleration, and case fatality rates relative to cases. They can objectively assess improvements of implemented epidemiologic measures and have applicable public health safety implications.

Novel uses of three-parameter logistic models and first-derivative models for the Coronavirus Disease (COVID-19) epidemic in the United States, in three distinct scenarios

Background: Forecasting the current coronavirus disease (COVID-19) epidemic in the United States necessitates novel mathematical models for accurate predictions. This paper examines novel uses of three-parameter logistic models and first-derivative models through three distinct scenarios that have not been examined in the literature as of July 14, 2020.Methods: Using publicly available data, statistical software was used to conduct a non-linear least-squares estimate to generate a three-parameter logistic model, with a subsequently generated first-derivative model. In the first scenario a logistic model was used to examine the natural log of COVID-19 cases as the dependent variable (versus day number), on July 11 and May 1. Independent t-test analyses were used to test comparative coefficient differences across models. In the second scenario, a first-derivative model was derived from a base three-parameter logistic model for April 27, examining time to peak mortality and decrease in case fatality rate. In the third scenario, a first-derivative model of mortality through July 11 as the dependent variable, versus confirmed cases, was generated to look at case fatality rate relative to increasing cases.Results: All models generated were statistically significant with R2 > 99%. The logistic models in the first scenario best predicted time to growth deceleration in the natural log of cases in the U.S. (slowing of exponential growth), estimated at March 11, 2020. For the May 1 data, independent t-test analyses of comparative coefficients across models were useful to track improvements from implemented public health measures. The first-derivative model in the second scenario on April 27, when the epidemic was more controlled, showed peak mortality around April 12-13, with a case fatality rate of < 1,000 deaths and trending down. The first-derivative model in the third scenario estimated a near-zero case fatality rate to occur at 4 million confirmed cases. It has not been affected by fluctuations in mortality from June 29 through July 11.Conclusion: Three-parameter logistic models and first-derivative models have utility in predicting time to growth deceleration, and case fatality rates relative to cases. They can objectively assess improvements of implemented epidemiologic measures and have applicable public health safety implications.

Novel uses of three-parameter logistic models and first-derivative models for the Coronavirus Disease (COVID-19) epidemic in the United States, in three distinct scenarios

Background Forecasting the current coronavirus disease (COVID-19) epidemic in the United States necessitates novel mathematical models for accurate predictions. This paper examines novel uses of three-parameter logistic models and first-derivative models through three distinct scenarios that have not been examined in the literature as of July 14, 2020. Methods Using publicly available data, statistical software was used to conduct a non-linear least-squares estimate to generate a three-parameter logistic model, with a subsequently generated first-derivative model. In the first scenario a logistic model was used to examine the natural log of COVID-19 cases as the dependent variable (versus day number), on July 11 and May 1. Independent t-test analyses were used to test comparative coefficient differences across models. In the second scenario, a first-derivative model was derived from a base three-parameter logistic model for April 27, examining time to peak mortality and decrease in case fatality rate. In the third scenario, a first-derivative model of mortality through July 11 as the dependent variable, versus confirmed cases, was generated to look at case fatality rate relative to increasing cases. Results All models generated were statistically significant with R2 > 99%. The logistic models in the first scenario best predicted time to growth deceleration in the natural log of cases in the U.S. (slowing of exponential growth), estimated at March 11, 2020. For the May 1 data, independent t-test analyses of comparative coefficients across models were useful to track improvements from implemented public health measures. The first-derivative model in the second scenario on April 27, when the epidemic was more controlled, showed peak mortality around April 12-13, with a case fatality rate of < 1,000 deaths and trending down. The first-derivative model in the third scenario estimated a near-zero case fatality rate to occur at 4 million confirmed cases. It has not been affected by fluctuations in mortality from June 29 through July 11. Conclusion Three-parameter logistic models and first-derivative models have utility in predicting time to growth deceleration, and case fatality rates relative to cases. They can objectively assess improvements of implemented epidemiologic measures and have applicable public health safety implications.