A Review for Neuroimaging Techniques in Multimedia Learning: Methodological Framework

Introduction

Cognitive load theory (CLT) aims to explain the effects of cognitive processing load due to learning tasks during new cognitive processing and information processing in long-term memory (Sweller et al., 2019). The heart of the CLT depends on the necessity of processing information in working memory before being stored in long-term memory in an information processing model (Anmarkrud et al., 2019). But, working memory which processes limited information at a time, limits information processing (Sweller, 2010). For this reason, the instructional design must decrease the unnecessary cognitive load in working memory. Otherwise, if working memory capacity exceeds and cognitive overload occurs, it may cause learning to be obscured (Leppink et al., 2013).

According to CLT, three types of cognitive load affect working memory during learning tasks; intrinsic cognitive load, extraneous cognitive load, and germane cognitive load (Paas & Sweller, 2014, p. 37). Intrinsic cognitive load refers to the inherent difficulty and complexity of every learning task. Extraneous load deals with when the learner has to process unnecessary interactive elements due to inappropriate instructional design. Germane cognitive load refers to the load occurring in working memory by learning, as when relating information between long-term memory and new information construction process. On the other hand, in their recent update of cognitive load theory, Sweller et al. (2011) preferred the term “germane resources'' because, unlike intrinsic and extraneous cognitive load, the germane load is not imposed by nature or structure of the learning material.

Measuring cognitive load is very important to provide guidelines and implications for successful instructional design (Korbach et al., 2017). Because implications for CL research allow determining possibility levels of CL in the early design phase (Paas et al., 2003). CL is measured with two different instruments, subjective measures for CL and objective measures for CL. Neuroimaging is an objective and direct type of measure for CL (Chang et al., 2016; A. Dan & Reiner, 2017; M. Mazher et al., 2017).

In recent years, a range of highly sensitive scanning technologies has been developed for computer imaging researchers to monitor human brain processing. By benefiting from the literature, current brain imaging techniques are grouped into three types:

Structural; Computerized Tomography (CT) (Schlorhaufer et al., 2012), Magnetic Resonance Imaging (MRI) (Juanes et al., 2015)

Functional; Electroencephalography (EEG) (Castro-Meneses et al., 2020), Positron emission tomography (PET) (Ruisoto Palomera et al., 2014), Functional magnetic resonance imaging (fMRI) (C. Liu et al., 2020), Functional near-infrared spectroscopy(fNIRS) (Uysal, 2016), Magnetoencephalography (MEG) (C.-J. Liu & Chiang, 2014), Infrared Thermography (IT) (Jenkins & Brown, 2014), Transcranial Magnetic Stimulation (TMS) (Hilbert et al., 2019) and, Transcranial Doppler Ultrasonography (TCD) (J. J. Loftus et al., 2018).

Metabolic; Magnetic Resonance Spectroscopy (MRS) (Ross & Sachdev, 2004).

Neuroimaging techniques may provide opportunities to measure CL and different types of CL accurately (Whelan, 2007) because fMRI and PET techniques use blood flow to collect data for CL during brain activity (Antonenko et al., 2010). fMRI can be effective in measuring intrinsic CL (Whelan, 2007). An alternative device to fMRI is fNIRS due to its smaller size to measure brain activity while collecting cortical blood flow data (Antonenko et al., 2010). EEG collects data as electrical activity occurs in brain regions (Jan-Louis Kruger & Doherty, 2016). Studies measure extraneous CL with EEG in the literature (e.g., Dan & Reiner, 2017; Örün & Akbulut, 2019).