Understanding underscore's implementation of isNaN

Question

Taken from the underscore.js source:

_.isNaN = function(obj) {
  return _.isNumber(obj) && obj != +obj;
};

Why did they do it this way? Is the above implementation equivalent to:

_.isNaN = function(obj) {
  return obj !== obj;
};

If it is, why the "more complicated" version? If it is not, what are the behavioural differences?

Answer 1

_.isNaN(new Number(NaN)) returns true.

And that's by design.

var n = new Number(NaN);
console.log(_.isNaN(n), n!==n); // logs true, false