摘要目的:探究以游戏为基础促进交流与行为的干预(play-based communication and behavior intervention，PCBI)对孤独症谱系障碍(autism spectrum disorder，ASD)幼儿内向性和外向性行为的影响。方法:2018年11月至2021年11月，选取90例在南京医科大学附属脑科医院儿童心理卫生研究中心登记进行PCBI超早期干预训练的8~30月龄ASD幼儿，按照2∶1的比例随机分配到干预组( n＝60)和等候组( n＝30)。采集12周干预前后幼儿与照顾者自由游戏的行为学录像，应用行为观察分析系统(Observer XT)对亲子互动状态和幼儿内外向性行为进行编码。采用SPSS 23.0统计软件对数据进行 t检验、卡方检验、Pearson相关分析和多元线性逐步回归分析。 结果:干预后干预组幼儿内外向性行为[(2.43±2.22)分，(1.88±1.91)分]较干预前[(4.82±3.37)分，(3.68±5.68)分]均有改善，均差异有统计学意义( t＝4.66，2.60，均 P<0.05)，等候组干预前[(4.23±2.47)分，(4.00±2.18)分]和干预后[(4.37±2.57)分，(4.67±3.72)分]均差异无统计学意义( t＝－0.23，－0.83，均 P>0.05)。干预后，干预组幼儿内外向性行为较等候组减少，均差异有统计学意义( t＝－3.70，－4.71，均 P<0.05)。两组干预前后内向性行为差值[(2.38±3.96)分，(1.80±5.37)分]和外向性行为差值[(1.80±5.37)分，(－0.67±4.38)分]均差异有统计学意义( t＝3.03，2.18，均 P<0.05)。多元线性逐步回归结果表明照顾者消极互动( β＝0.29， P<0.01)是内向性行为的危险因素；儿童消极互动( β＝0.45， P<0.01)是外向性行为的危险因素，儿童物体状态( β＝－0.30， P<0.01)和二元互动( β＝－0.30， P<0.01)是外向性行为保护因素。干预组ASD幼儿的亲子互动状态在干预前后儿童积极互动、照顾者积极互动和二元互动增加( t＝－6.77，－4.58，－7.72，均 P<0.05)，儿童物体状态和照顾者消极互动均减少( t＝3.37，4.30，均 P<0.05)。 结论:PCBI超早期干预可以有效改善ASD幼儿的内外向性行为。内向性行为的改善可能是通过减少照顾者消极互动起作用的，外向性行为的改善可能是通过增加亲子二元互动起作用的。

abstractsObjective:To explore the effect of play-based communication and behavior intervention (PCBI) on internalizing and externalizing behaviors in children with autism spectrum disorder (ASD).Methods:From November 2018 to November 2021, 90 ASD children aged 8-30 months who registered in the Child Mental Health Research Center of the Affiliated Brain Hospital of Nanjing Medical University for PCBI ultra-early intervention training were selected and randomly assigned to the intervention group( n=60) and the waiting group( n=30) according to the ratio of 2∶1.Behavioral videos of free play between children and caregivers were collected before and after the 12-week intervention and the behavioral observation and analysis system (Observer XT) was used to code the parent-child interaction status and children's internalizing and externalizing behaviors.SPSS 23.0 statistical software were used and the data were analyzed by t-test, chi-square test, Pearson correlation analysis and multiple linear stepwise regression analysis. Results:After the intervention, the children in the intervention group had improved internalizing and externalizing behaviors (2.43±2.22, 1.88±1.91) compared with those before the intervention (4.82±3.37, 3.68±5.68), and the difference was statistically significant ( t=4.66, 2.60, both P<0.05). The children in the waiting group had no significant difference in internalizing and externalizing behaviors before (4.23±2.47, 4.00±2.18) and after intervention (4.37±2.57, 4.67±3.72) ( t=-0.23, -0.83, both P>0.05). After intervention, the children in the intervention group had fewer internalizing and externalizing behaviors than those in the waiting group, and the difference was statistically significant ( t=-3.70, -4.71, both P<0.05). The differences in internalizing behaviors (2.38±3.96, 1.80±5.37) and externalizing behaviors (1.80±5.37, -0.67±4.38) between the two groups before and after the intervention were statistically significant ( t=3.03, 2.18, both P<0.05). The results of multiple linear stepwise regression showed that the negative interaction of caregivers ( β=0.29, P<0.01) was a risk factor for internalizing behaviors.The negative interaction of children ( β=0.45, P<0.01) was a risk factor for externalizing behaviors, and the child's object status ( β=-0.30, P<0.01) and binary interaction ( β=-0.39, P<0.01) were protective factors for externalizing behaviors.In the intervention group, active child interaction, active caregiver interaction, and binary interaction increased after the intervention ( t=-6.77, -4.58, -7.72, all P<0.05), while the child's object status and the caregiver's negative interaction decreased ( t=3.37, 4.30, both P<0.05). Conclusion:The PCBI ultra-early intervention can effectively improve the internalizing and externalizing behaviors of ASD children.Improvements in internalizing behaviors may work by reducing negative caregiver interactions, and improvements in externalizing behaviors may work by increasing parent-child binary interactions.