Representational similarity analysis (RSA) is a multivariate technique to investigate cortical representations of objects or constructs. While avoiding ill-posed matrix inversions that plague multivariate approaches in the presence of many outcome variables, it suffers from the confound arising from the non-orthogonality of the design matrix. Here, a partial correlation approach will be explored to adjust for this source of bias by partialling out this confound in the context of the searchlight method for functional imaging datasets. A formal analysis will show the existence of a dependency of this confound on the temporal correlation model of the sequential observations, motivating a data-driven approach that avoids the problem of misspecification of this model. However, where the autocorrelation locally diverges from its volume estimate, bias may be difficult to control for exactly, given the difficulties of estimating the precise form of the confound at each voxel. Application to real data shows the effectiveness of the partial correlation approach, suggesting the impact of local bias to be minor. However, where the control for bias locally fails, possible spurious associations with the similarity matrix of the stimuli may emerge. This limitation may be intrinsic to RSA applied to non-orthogonal designs. The software implementing the approach is made publicly available (https://github.com/roberto-viviani/rsa-rsm.git).
翻译:代表相似性分析(RSA)是一种多变的方法,用于调查物体或构造的圆形表达方式。虽然避免了在存在许多结果变量的情况下使多变方法困扰多变方法的不正确矩阵反转,但它受到设计矩阵非横向性引起的混乱的困扰。在这里,将探索一种部分关联性方法,通过在功能成像数据集的探照灯方法中将这种混为一体的方法来调整这种偏差的来源。一种正式分析将显示这种混为在连续观测的时间相关模型上的存在,同时鼓励采用一种数据驱动方法,避免出现该模型的误差。然而,如果当地离异性与设计表的数值不同,则可能难以准确控制偏差,因为难以估计每个 voxel的混杂的确切形式。对真实数据的应用表明部分关联方法的有效性,表明局部偏差的影响不大。然而,如果对地方偏差的控制失败,则可能与类似S&VI/Complia 的内向性设计模型存在误差性关联。