It is typical to assign command results to a variable. What does this variable contain if a command outputs nothing? It turns out that in PowerShell 3.0+ it is something that looks like null and behaves like null in most cases except some special. Such special cases are potential traps.
The script Test-1-different-results.ps1 shows that "no results" assigned to a variable looks like null but may cause different results, comparing with a true null variable.
The script Test-2-not-exactly-null.ps1 shows how "no results" null can be distinguished from a true null. This is just a fact, not a suggested method. This feature is not documented in known official sources.
Note: yet another difference between PowerShell 2.0 and 3.0 was found. In the
strict mode $null.PSObject fails in v2 and works in v3+. In the latter case
it gets null for a true null and a PSCustomObject instance for "no results".