From the abstract: To investigate how play affects variations in language, parents and children were assigned to 1 of 3 conditions: free play with blocks, guided play, or play with preassembled structures (Study 1). Parents in the guided play condition produced significantly higher proportions of spatial talk than parents in the other two conditions, and children in the guided play condition produced significantly more spatial talk than those in the free play condition.
From this we can pick up that parents use more spatial language in guided play with blocks, and kids either pick up on it or desire to express those thoughts themselves. The authors conclude that this finding suggests "simple-to-execute educational interventions." Since spatial conceptual ability has been tied to math and science performance, this is something we want to encourage at an early age.
Cool, so if you want your kids to develop spatial language and presumably spatial attention skills, make sure they have guided play with Lego rather than free play, right?
But is that a valid conclusion? It seems to me the study shows that guided block play probably helps develop spatial concepts in the context of normal development, but there is nothing here that shows that children with sub-par ability benefit from such activities.
Another 2011 study referenced in the same article is Children’s spatial thinking: does talk about the spatial world matter? by Pruden, Levin and Huttenlocher in Developmental Science, looks at " the relations between parent spatial language input, children’s own production of spatial language, and children’s later spatial abilities." This longitudinal study backs up the inference that children pick up spatial talk from their parents in the first few years of life, and shows that this exposure leads to improved performance on spatial problem solving tasks later on (4 1/2 years old).
This again supports the notion that you should a) play with your kids and b) use spatial language (words such as big, tall, circle, curvy, edge, etc.) with them when you do, whether it's building with blocks or playing catch. Once again, however, these results are in the context of normal development, not educational intervention, and we should not make blanket suppositions about transference.
However, with other studies showing that university students can improve spatial selective attention by playing an action video game for 10 hours [1], or that such games increase visual processing speed [2] and can significantly improve reading speed in children with dyslexia [3], it does become reasonable to suppose it couldn't hurt.
References 1. Green & Bavelier (2003). Action video game modifies visual selective attention. Nature, 423, 534-537.
2. Dye, Green & Bavelier (2009). Increasing speed of processing with action video games. Current Directions in Psychological Science, 18(6), 321-326.
3. Franceschini et al. (2013), Action video games make dyslexic children read better. Current Biology, 23, 1-5.