Such plots can be uninformative and even misleading, I would recommend partial residual plots, to additionally examine the partial relation that is the conditional relation of y on one X taking account of the other X’s .
Such a plot is often called a response surface plot and is used in response surface analysis in chemistry and chemical engineering. See the chapter in this reference: https://www.google.com/search?q=montgomery+douglas+c.+design+and+analysis+of+experiments&oq=Montgomery&aqs=chrome.1.69i57j69i59j0l5j69i60.9978j0j7&sourceid=chrome&ie=UTF-8
The plots are often done in R, see: https://www.google.com/search?ei=uHFHXsanPMqMtAayiraYCg&q=response+surface+methodology+plots+r&oq=Response+surface+plots&gs_l=psy-ab.1.5.0j0i22i30l8.1945798.1968916..1977923...0.2..3.152.5446.63j3......0....1..gws-wiz.....6..0i71j0i362i308i154i357j0i273j0i131j0i67j0i131i67j0i10.L_H5A69o1ak
I don't think this is currently in SPSS See: https://www.google.com/search?ei=dHlHXo6AL8qxtQaE1r2YDw&q=response+surface+methodology+plots+SPSS&oq=response+surface+methodology+plots+SPSS&gs_l=psy-ab.3..33i22i29i30.128319.132350..138930...0.2..0.123.459.4j1......0....1..gws-wiz.......0i71j0i22i30.cQEfms5rhrU&ved=0ahUKEwiOipjo4dLnAhXKWM0KHQRrD_MQ4dUDCAs&uact=5 As Kelvyn Jones indicates such plots can be cofusing depending on the number and kinds of predictors in your particulat application.