<p>In the digital age, Augmented Reality (AR) technology, an interactive medium that superimposes virtual information onto the physical world, has expanded its application scope in educational settings in recent years. While AR demonstrates immense potential for educational applications, its impact on learners’ cognitive load remains a subject of ongoing debate. In this study, we used meta-analysis to systematically synthesize the findings of 27 experimental and quasi-experimental studies conducted in China and internationally and explored the effects of AR technology on learners’ cognitive load. The findings showed that AR technology has a small-to-moderate significant effect on reducing learners’ cognitive load (Hedges’ <InlineEquation ID="IEq1"> <EquationSource Format="TEX">$\textit{g} = -0.383$</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi mathvariant="italic">g</mi> <mo>=</mo> <mo>-</mo> <mn>0.383</mn> </mrow> </math></EquationSource> </InlineEquation>, 95% CI [<InlineEquation ID="IEq2"> <EquationSource Format="TEX">$-0.740, -0.027$</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mo>-</mo> <mn>0.740</mn> <mo>,</mo> <mo>-</mo> <mn>0.027</mn> </mrow> </math></EquationSource> </InlineEquation>], <InlineEquation ID="IEq3"> <EquationSource Format="TEX">$\textit{p} = 0.035$</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi mathvariant="italic">p</mi> <mo>=</mo> <mn>0.035</mn> </mrow> </math></EquationSource> </InlineEquation>), with extremely high heterogeneity across studies (<InlineEquation ID="IEq4"> <EquationSource Format="TEX">$I^2 = 92.7\%$</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <msup> <mi>I</mi> <mn>2</mn> </msup> <mo>=</mo> <mn>92.7</mn> <mo>%</mo> </mrow> </math></EquationSource> </InlineEquation>, <InlineEquation ID="IEq5"> <EquationSource Format="TEX">$p < 0.001$</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>p</mi> <mo><</mo> <mn>0.001</mn> </mrow> </math></EquationSource> </InlineEquation>). Publication bias tests indicated borderline small-study effects (Egger’s test <InlineEquation ID="IEq6"> <EquationSource Format="TEX">$\textit{p} = 0.072$</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi mathvariant="italic">p</mi> <mo>=</mo> <mn>0.072</mn> </mrow> </math></EquationSource> </InlineEquation>), but sensitivity analysis (trim-and-fill method) confirmed the robustness of the pooled effect (adjusted <InlineEquation ID="IEq7"> <EquationSource Format="TEX">$\textit{g} = -0.446$</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi mathvariant="italic">g</mi> <mo>=</mo> <mo>-</mo> <mn>0.446</mn> </mrow> </math></EquationSource> </InlineEquation>, 95% CI [<InlineEquation ID="IEq8"> <EquationSource Format="TEX">$-0.804, -0.087$</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mo>-</mo> <mn>0.804</mn> <mo>,</mo> <mo>-</mo> <mn>0.087</mn> </mrow> </math></EquationSource> </InlineEquation>]). Moderating effect analysis revealed that the frequency of intervention (<InlineEquation ID="IEq9"> <EquationSource Format="TEX">$\textit{Q}=10.369, \textit{p}=0.006$</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi mathvariant="italic">Q</mi> <mo>=</mo> <mn>10.369</mn> <mo>,</mo> <mi mathvariant="italic">p</mi> <mo>=</mo> <mn>0.006</mn> </mrow> </math></EquationSource> </InlineEquation>) and the form of feedback (<InlineEquation ID="IEq10"> <EquationSource Format="TEX">$\textit{Q}=13.245, \textit{p}=0.004$</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi mathvariant="italic">Q</mi> <mo>=</mo> <mn>13.245</mn> <mo>,</mo> <mi mathvariant="italic">p</mi> <mo>=</mo> <mn>0.004</mn> </mrow> </math></EquationSource> </InlineEquation>) had a significant moderating effect on the effect of AR technology in reducing cognitive load. In contrast, the moderating effect of AR technology application disciplines, educational stages, intervention time, learning styles, device types and cognitive load measurement tool on the effect of AR technology in reducing cognitive load was not significant. Therefore, the study suggests that educational practices should promote the effective application of AR technology in education by controlling the intervention tempo, optimizing the feedback mechanism, and adapting the content design to the characteristics of disciplines and the educational stage.</p>

Can augmented reality technology reduce learners' cognitive load? A meta-analysis

期刊论文
出版模式：

开放获取
发表日期：

2026年1月26日

查看全文

Xiuqi Zhu,
Kuan Peng,
Shuyu Yu,
Guohua Wang

摘要

In the digital age, Augmented Reality (AR) technology, an interactive medium that superimposes virtual information onto the physical world, has expanded its application scope in educational settings in recent years. While AR demonstrates immense potential for educational applications, its impact on learners’ cognitive load remains a subject of ongoing debate. In this study, we used meta-analysis to systematically synthesize the findings of 27 experimental and quasi-experimental studies conducted in China and internationally and explored the effects of AR technology on learners’ cognitive load. The findings showed that AR technology has a small-to-moderate significant effect on reducing learners’ cognitive load (Hedges’ $\textit{g} = -0.383$ $g = - 0.383$ , 95% CI [ $-0.740, -0.027$ $- 0.740, - 0.027$ ], $\textit{p} = 0.035$ $p = 0.035$ ), with extremely high heterogeneity across studies ( $I^2 = 92.7\%$ $I^{2} = 92.7 %$ , $p < 0.001$ $p < 0.001$ ). Publication bias tests indicated borderline small-study effects (Egger’s test $\textit{p} = 0.072$ $p = 0.072$ ), but sensitivity analysis (trim-and-fill method) confirmed the robustness of the pooled effect (adjusted $\textit{g} = -0.446$ $g = - 0.446$ , 95% CI [ $-0.804, -0.087$ $- 0.804, - 0.087$ ]). Moderating effect analysis revealed that the frequency of intervention ( $\textit{Q}=10.369, \textit{p}=0.006$ $Q = 10.369, p = 0.006$ ) and the form of feedback ( $\textit{Q}=13.245, \textit{p}=0.004$ $Q = 13.245, p = 0.004$ ) had a significant moderating effect on the effect of AR technology in reducing cognitive load. In contrast, the moderating effect of AR technology application disciplines, educational stages, intervention time, learning styles, device types and cognitive load measurement tool on the effect of AR technology in reducing cognitive load was not significant. Therefore, the study suggests that educational practices should promote the effective application of AR technology in education by controlling the intervention tempo, optimizing the feedback mechanism, and adapting the content design to the characteristics of disciplines and the educational stage.