人-智能系统认知适配对人机协同决策绩效的影响

doi:10.16265/j.cnki.issn1003-3033.2026.08.0813

摘要/Abstract

摘要：

为解决煤矿智能监控任务中作业人员与智能系统认知差异引致的人机协同决策绩效受限问题,基于认知负荷理论,以人-智能系统认知适配为切入点,引入任务复杂性(信息负荷、任务结构)与系统透明度(信息透明度、过程透明度)作为情境调节变量,构建认知适配影响人机协同决策绩效的理论模型;采用多阶段问卷调查方法,实证分析煤矿一线作业人员数据。结果表明:“人-智能系统”认知适配显著正向影响人机协同决策绩效(标准化回归系数β=0.253,显著性指标p<0.001);信息负荷、任务结构、信息透明度和过程透明度均显著正向调节认知适配与人机协同决策绩效之间的关系,即在相应水平较高时,认知适配对协同决策绩效的促进作用更强。其中,高信息负荷和高过程透明度情境下,该促进作用尤为显著。

关键词: 人-智能系统, 认知适配, 人机协同决策绩效, 任务复杂性, 系统透明度

Abstract:

To address constrained human-machine collaborative decision-making performance caused by cognitive differences between operators and intelligent systems in coal mine intelligent monitoring tasks, this study, grounded in cognitive load theory, develops a theoretical model of the effects of human-intelligent system cognitive fit on human-machine collaborative decision-making performance. Task complexity, including information load and task structure, and system transparency, including information transparency and process transparency, are introduced as contextual moderators. Using a multi-stage questionnaire survey, empirical data were collected from frontline coal mine workers. The results show that human-intelligent system cognitive fit has a significant positive effect on human-machine collaborative decision-making performance (β=0.253, p<0.001). Information load, task structure, information transparency, and process transparency all significantly and positively moderate the relationship between cognitive fit and human-machine collaborative decision-making performance. Specifically, higher levels of these contextual factors strengthen the positive effect of cognitive fit on collaborative decision-making performance. This effect is particularly pronounced under conditions of high information load and high process transparency.

Key words: human-intelligent system, cognitive fit, human-machine collaborative decision-making performance, task complexity, system transparency

中图分类号:

X912.9安全人机学

牛莉霞, 李波, 李国, 潘力航, 张国建. 人-智能系统认知适配对人机协同决策绩效的影响[J]. 中国安全科学学报, 2026, 36(8): 21-28.

Niu Lixia, Li Bo, Li Guo, Pan Lihang, Zhang Guojian. Effect of human-intelligent system cognitive fit on human-machine collaborative decision-making performance[J]. China Safety Science Journal, 2026, 36(8): 21-28.

图/表 6

图1

表1

表2

表3

图2

表4

参考文献 33

[1]	王大龙, 王冰山, 曹睿, 等. 煤矿安全管理行为交互机制与模型[J]. 中国安全科学学报, 2025, 35(5): 32-38. doi: 10.16265/j.cnki.issn1003-3033.2025.05.1094
	Wang Dalong, Wang Bingshan, Cao Rui, et al. Interaction mechanism and model of coal mine safety management behaviors[J]. China Safety Science Journal, 2025, 35(5): 32-38. doi: 10.16265/j.cnki.issn1003-3033.2025.05.1094
[2]	牛莉霞, 李肖萌. 5G时代智慧矿山安全管理新模式[J]. 中国安全科学学报, 2021, 31(6):29-36. doi: 10.16265/j.cnki.issn 1003-3033.2021.06.004
	Niu Lixia, Li Xiaomeng. New safety management mode of smart mine in the 5G era[J]. China Safety Science Journal, 2021, 31(6): 29-36. doi: 10.16265/j.cnki.issn 1003-3033.2021.06.004
[3]	张拴民, 史洪恺. 基于5G技术的智能化煤矿安全管控平台设计[J]. 中国安全科学学报, 2024, 34(增刊1):140-147.
	Zhang Shuanmin, Shi Hongkai. Design of intelligent coal mine safety management and control platform based on 5G technology[J]. China Safety Science Journal, 2024, 34(S1): 140-147. doi: 10.16265/j.cnki.issn1003-3033.2024.S1.0023
[4]	许为, 高在峰, 葛列众. 智能时代人因科学研究的新范式取向及重点[J]. 心理学报, 2024, 56(3): 363-382. doi: 10.3724/SP.J.1041.2024.00363
	Xu Wei, Gao Zaifeng, Ge Liezhong. New paradigm orientations and key issues of human factors research in the intelligent era[J]. Acta Psychologica Sinica, 2024, 56(3): 363-382. doi: 10.3724/SP.J.1041.2024.00363
[5]	Robey D, Taggart W. Human information processing in information and decision support systems[J]. MIS Quarterly, 1982, 6(2):61-77.
[6]	梁雅婷. “人-智能体”认知适配量表的编制及应用[D]. 杭州: 浙江大学, 2023.
	Liang Yating. Development and application of the human-agent cognitive adaptation scale[D]. Hangzhou: Zhejiang University, 2023.
[7]	Parasuraman R, Sheridan T B, Wickens C D. Situation awareness, mental workload, and trust in automation: viable, empirically supported cognitive engineering constructs[J]. Journal of Cognitive Engineering and Decision Making, 2008, 2(2):140-160. doi: 10.1518/155534308X284417
[8]	De Visser E J, Pak R, Shaw T H. From 'automation' to 'autonomy': the importance of trust repair in human-machine interaction[J]. Ergonomics, 2018, 61(10):1409-1427. doi: 10.1080/00140139.2018.1457725 pmid: 29578376
[9]	刘记海. 人工智能信任对人机协同决策绩效的影响研究:人机知识转移与知识整合视角[D] 杭州: 杭州电子科技大学, 2024.
	Liu Jihai. Research on the influence of trust in artificial intelligence on human-machine collaborative decision-making performance: perspectives of human-machine knowledge transfer and knowledge integration[D]. Hangzhou: Hangzhou Dianzi University, 2024.
[10]	Sweller J. Element interactivity and intrinsic, extraneous, and germane cognitive load[J]. Educational Psychology Review, 2010, 22(2):123-138. doi: 10.1007/s10648-010-9128-5
[11]	Scheutz M, DeLoach S A, Adams J. A framework for developing and using shared mental models in human-agent teams[J]. Journal of Cognitive Engineering and Decision Making, 2017, 11(3):203-224. doi: 10.1177/1555343416682891
[12]	Fan Xiaocong, Yen J. Modeling cognitive loads for evolving shared mental models in human-agent collaboration[J]. IEEE Transactions on Systems, Man, and Cybernetics-Part B: Cybernetics, 2011, 41(2):354-367. doi: 10.1109/TSMCB.2010.2053705
[13]	高锦萍, 李沛怡. 会计任务复杂性、可信性感知与人机协作模式选择[J]. 财经论丛, 2024(11):80-91.
	Gao Jinping, Li Peiyi. Accounting task complexity, perceived credibility and selection of human-machine collaboration modes[J]. Collected Essays on Finance and Economics, 2024(11): 80-91.
[14]	Campbell D J. Task complexity: a review and analysis[J]. Academy of Management Review, 1988, 13(1):40-52. doi: 10.2307/258353
[15]	Eppler M J, Mengis J. The concept of information overload: a review of the literature from organization science, accounting, marketing, MIS, and related disciplines[J]. The Information Society, 2004, 20(5):325-344. doi: 10.1080/01972240490507974
[16]	Schick A G, Gordon L A, Haka S. Information overload: a temporal approach[J]. Accounting, Organizations and Society, 1990, 15(3):199-220. doi: 10.1016/0361-3682(90)90005-F
[17]	Campbell D J. Task complexity: a review and analysis[J]. The Academy of Management Review, 1988, 13(1):40-52. doi: 10.2307/258353
[18]	Wood R E. Task complexity: definition of the construct[J]. Organizational Behavior and Human Decision Processes, 1986, 37(1):60-82. doi: 10.1016/0749-5978(86)90044-0
[19]	李浩, 郭春红, 宋芳柏. AI自动化对认知重塑的双刃剑效应研究[J]. 管理学报, 2025, 22(2):345-354.
	Li Hao, Guo Chunhong, Song Fangbai. Research on the double-edged sword effect of AI automation on cognitive reshaping[J]. Chinese Journal of Management, 2025, 22(2): 345-354.
[20]	Sweller J. Cognitive load during problem solving: effects on learning[J]. Cognitive Science, 1988, 12(2):257-285. doi: 10.1207/s15516709cog1202_4
[21]	Fügener A, Grahl J, Gupta A, et al. Cognitive challenges in human-artificial intelligence collaboration: investigating the path toward productive delegation[J]. Information Systems Research, 2022, 33(2):678-696. doi: 10.1287/isre.2021.1079
[22]	Seeber I, Bittner E, Briggs R O, et al. Machines as teammates: a research agenda on AI in team collaboration[J]. Information & Management, 2020, 57(2):1-22.
[23]	Mercado J E, Rupp M A, Chen J Y C, et al. Intelligent agent transparency in human-agent teaming for multi-UxV management[J]. Human Factors, 2016, 58(3):401-415. doi: 10.1177/0018720815621206 pmid: 26867556
[24]	Sanders T L, Wixon T, Schafer K E, et al. The influence of modality and transparency on trust in human-robot interaction[C]. Proceedings of the 2014 IEEE International Inter-Disciplinary Conference on Cognitive Methods in Situation Awareness and Decision Support (CogSIMA). IEEE, 2014:156-159.
[25]	Vössing M, Kühl N, Lind M, et al. Designing transparency for effective human-AI collaboration[J]. Information Systems Frontiers, 2022, 24(3):877-895. doi: 10.1007/s10796-022-10284-3
[26]	Alonso V, De La P P. System transparency in shared autonomy: a mini review[J]. Frontiers in Neurorobotics, 2018, 12:DOI:10.3389/fnbot.2018.00083.
[27]	Shamim S, Zeng Jing, Shariq S M, et al. Role of big data management in enhancing big data decision-making capability and quality among Chinese firms: a dynamic capabilities view[J]. Information & Management, 2019, 56(6):DOI:10.1016/j.im.2018.12.003.
[28]	Kuhn T, Poole M S. Do conflict management styles affect group decision making?[J]. Human Communication Research, 2000, 26(4):558-590. doi: 10.1111/j.1468-2958.2000.tb00769.x
[29]	Che Tao, Wu Zhengxin, Wang Yong, et al. Impacts of knowledge sourcing on employee innovation: the moderating effect of information transparency[J]. Journal of Knowledge Management, 2019, 23(2):221-239. doi: 10.1108/JKM-11-2017-0554
[30]	Durcikova A, Gray P H. How knowledge validation processes affect knowledge contribution[J]. Journal of Management Information Systems, 2009, 25(4):81-107. doi: 10.2753/MIS0742-1222250403
[31]	Li Jianping, Pan J S. Dimensions of artificial intelligence anxiety based on the integrated fear acquisition theory[J]. Technology in Society, 2020, 63: DOI:10.1016/j.techsoc.2020.101410.
[32]	Pope-Davis D B, Twing J S. The effects of age, gender, and experience on measures of attitude regarding computers[J]. Computers in Human Behavior, 1991, 7(4):333-339. doi: 10.1016/0747-5632(91)90020-2
[33]	Margrett J A, Marsiske M. Gender differences in older adults' everyday cognitive collaboration[J]. International Journal of Behavioral Development, 2002, 26(1):45-59. doi: 10.1080/01650250143000319 pmid: 20657668

模型	χ²	df	χ²/v	CFI	TLI	RMSEA	SRMR
6因子(A,B,C,D,E,F)	415.444	391	1.063	0.996	0.995	0.011	0.0721
5因子(A+B,C,D,E,F)	832.83	395	2.108	0.929	0.922	0.052	0.0569
4因子(A+B,C+D,E,F)	1 529.406	399	3.833	0.818	0.801	0.083	0.1257
3因子(A+B,C+D,E+F)	1 814.689	402	4.514	0.772	0.753	0.092	0.1358
2因子(A+B+C+D,E+F)	3 008.348	404	7.446	0.58	0.548	0.125	0.1292
1因子(A+B+C+D+E+F)	3 928.319	405	9.7	0.432	0.389	0.145	0.1407

变量	均值	标准差	相关系数
变量	均值	标准差	1	2	3	4	5	6	7	8	9	10	11
性别	1.14	0.35	1	—	—	—	—	—	—	—	—	—	—
年龄	2.76	1.15	0.162^**	1	—	—	—	—	—	—	—	—	—
工龄	2.98	1.29	0.102^*	0.879^**	1	—	—	—	—	—	—	—	—
学历	1.70	0.74	0.230^**	0.042	0.023	1	—	—	—	—	—	—	—
岗位类型	2.17	1.17	0.361^**	0.073	0.031	0.436^**	1	—	—	—	—	—	—
A	3.83	0.75	0.047	-0.029	-0.016	0.085	-0.019	1	—	—	—	—	—
F	3.71	0.87	-0.028	-0.068	-0.069	0.009	-0.005	0.271^**	1	—	—	—	—
B	2.36	1.09	-0.127	0.052	0.080	-0.066	-0.042	0.305	-0.19	1	—	—	—
C	2.38	1.09	-0.128	0.077	0.089	-0.073	-0.041	0.266	-0.176	0.655^**	1	—	—
D	1.94	0.88	0.005	0.050	0.072	-0.052	-0.032	-0.352	-0.298	0.289^**	0.265^**	1	—
E	2.01	0.91	-0.045	0.031	0.052	-0.127^**	-0.053	-0.303	-0.278	0.241^**	0.214^**	0.696^**	1

变量类型	变量	人机协同决策绩效
变量类型	变量	模型1	模型2	模型3	模型4	模型5
控制变量	性别	-0.047	-0.043	-0.055	-0.060	-0.057
	年龄	-0.044	-0.051	-0.043	-0.031	-0.043
	工龄	0.014	0.037	0.024	0.012	0.016
	学历	0.001	0.003	-0.002	-0.001	0.005
	岗位类型	-0.008	-0.036	-0.026	-0.007	-0.013
自变量	“人-智能系统”认知适配	0.253^***	0.122^**	0.157^**	0.207^***	0.192^***
调节变量	信息透明度	—	—	—	-0.087^*	—
	过程透明度	—	—	—	—	-0.072
	信息负荷	—	-0.142^***	—	—	—
	任务结构	—	—	-0.125^**	—	—
交互项	“人-智能系统”认知适配×信息负荷	—	0.218^***	—	—	—
	“人-智能系统”认知适配×任务结构	—	—	0.182^***	—	—
	“人-智能系统”认知适配×信息透明度	—	—	—	0.087^*	—
	“人-智能系统”认知适配×过程透明度	—	—	—	—	0.117^**
	决定系数R²	0.068	0.141	0.123	0.084	0.088
	决定系数变化量ΔR²		0.037	0.026	0.007	0.012
	F统计量F	4.887^***	8.289^***	7.067^***	4.625^***	4.844^***
	F值变化量ΔF	27.326^***	17.529^***	12.14^***	2.932^**	5.107^***

调节变量水平	效应值	标准误	95%置信区间
调节变量水平	效应值	标准误	低限	高限
低信息负荷	0.181 4	0.095	0.105 4	0.268 2
高信息负荷	0.414 0	0.076 5	0.263 5	0.564 5
低任务结构	0.182 8	0.092 6	0.119 2	0.284 8
高任务结构	0.379 4	0.071 8	0.038 3	0.520 6
低信息透明度	0.138 5	0.102 4	0.062 9	0.339 9
高信息透明度	0.392 1	0.090 2	0.214 7	0.569 5
低过程透明度	0.116 6	0.103 8	0.116 6	0.291 5
高过程透明度	0.418 2	0.088 5	0.244 2	0.592 1