ARY: Learning curve functions and formulas

In schmettow/asymptote: functions for learning curves

Description

exponential learning curve functions and non-linear regression formulas for use with brms

Usage

ARY

ary(ampl, rate, asym, trial)

SCOR

scor(scale, offset, rate, trial)

LSCOR

lscor(scale, offset, rate, trial)

LARY

lary(ampl, rate, trial)

iry(init, rate, asym, trial)

ERY

ery(pexp, rate, asym, trial)

LERY

lery(pexp, rate, asym, trial)

ARARY

arary(amplS, rateS, amplM, rateM, asym, trialS, trialM)

LARARY

larary(amplS, rateS, amplM, rateM, asym, trialS, trialM)

Arguments

ampl	amplitude (init - asym)
rate	rate of learning
asym	asymptote, max performance
trial	practice trial, starting at 0
init	initial performance
exp	previous experience

Format

An object of class formula of length 3.

An object of class formula of length 3.

An object of class formula of length 3.

An object of class formula of length 3.

An object of class formula of length 3.

An object of class formula of length 3.

An object of class formula of length 3.

An object of class formula of length 3.

Value

numeric

Computes the expected performance after practice trials using the exponential law of practice. The following parametrizations are provided:

ary

ampl, rate, asym

ery

pexp, rate, asym

iry

init, rate, asym

In addition, regression formulas are provided for the Bayesian regression engine brms. The right-hand side is always called performance. By convention, formulas use upper case names, whereas the functions use lower-case.

Furthermore, two-component learning functions are provided for cases where performance is composed of a general skill component S (e.g. learning to draw a figure using a mirror) and a motor sequence component M (e.g., learning to draw the exact same figure)

arary

ampl, rate, asym

All functions and formulas are also available with link functions that put the parameters on a linear predictor scale (supported range [-∞; ∞]) by log transformation. These formnulas/functions have the prefix L (e.g., LARA)

Author(s)

Martin Schmettow

Examples

ARY
ary(2, 0.3, 1, 1:5)
LARY
lary(log(2), log(0.3), log(1), 1:5)

schmettow/asymptote documentation built on Oct. 25, 2020, 10:16 p.m.