Development of visual noise mask for human point-light displays

• Main Author: Senra, Catarina Carvalho
• Publication Date: 2022
• Other Authors: Sampaio, Adriana, Lapenta, Olivia Morgan
• Language: eng
• Source: Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
• Download full: https://hdl.handle.net/1822/79502

Since the 1970’s Point-Light Display (PLD) stimuli have been widely used in investigations regarding how humans process and recognize human movement. Because humans have a great ability to recognize human motion even in the absence of pictorial information, several studies introduce visual noise in order to make stimuli recognition more challenging. The usual approach is to introduce extra moving dots of similar size that move along to the actual human PLD. To construct such a noise mask, often researchers must develop algorithms that generate random moving dots. Although some authors made platforms allowing manipulations within the algorithmic possibilities available, most of the developed and available ways of visual dot noise masks production rely on paid softwares, have file format restrictions and require the researcher to have extensive programming skills. In this regard, we herein propose to build the noise mask on Blender, a free open source software, with a graphical interface that reads and exports many file formats and enables the manipulation of videos both in 2D and 3D. Therefore, we present a user- friendly step-by-step guide on how to develop visual noise for masking PLD. Specifically, we explain how to set a dynamic movement in a 2D environment that relies mainly on changing an object position on the ‘x' and ‘y' axis. Additionally, we also present how to build the dots and how they can be manipulated to create the desired movement. The herein presented guideline can also be easily translated and applied in the 3D option. Furthermore, we made available the environment of the software with some directions and the set of noise videos developed by our group. Finally, as normally the combination of videos with the mask is crucial, the process of combination of the videos and the mask in Blender is also explained. In sum, the main advantages of the presented methodology are the non-expensiveness and no need of programming experience; thus having no prerequisites to be applied. In particular, this step-by-step guide might be appealing to students engaged in this research topic but who are still novice in programming skills usually required to build visual dot noise masks.