The simple reason is that a dog will pick up subtle cues of body langauge that humans (having verbal language) either do not notice, or have no way of consciously knowing about.  Dogs live in a world of non-verbal cues which they have to interpret correctly to both avoid trouble (with other dogs, especially) and to get food and shelter.

The first thing about dogs that has to be remembered is that they live in a hierarchical society.  They have to.  A lone dog, left alone, will soon go crazy.  And they have to be in a hierarchy and know their place in it.  

Humans, particularly dog handlers, can only "control" their dogs by placing themselves in the dominant position in that hierarchy.  A dog which understands a human to be in a submissive/subordinate position will take the dominant position and "walk all over" that person.  Obviously, a dog handler is not much good if he/she cannot control their dog, and that means they have to be in the dominant position.

A dog, when in a subordinate position, will do everything in its power to please the dog (or person) in the dominant position.  To accomplish this, the dog will learn the subtle cues of physical language - motion, posture, etc. - which indicate whether the dominant is pleased or not.  

So, these drug dogs will do what pleases their master.  They will knock themselves out to do it.

Official conduct that does not "compromise any legitimate interest in privacy" is not a search subject to the Fourth Amendment. Jacobsen, 466 U.S., at 123. We have held that any interest in possessing contraband cannot be deemed "legitimate," and thus, governmental conduct that only reveals the possession of contraband "compromises no legitimate privacy interest." Ibid. This is because the expectation "that certain facts will not come to the attention of the authorities" is not the same as an interest in "privacy that society is prepared to consider reasonable." Id., at 122 (punctuation omitted). In United States v. Place, 462 U.S. 696 (1983), we treated a canine sniff by a well-trained narcotics-detection dog as "sui generis" because it "discloses only the presence or absence of narcotics, a contraband item." Id., at 707; see also Indianapolis v. Edmond, 531 U.S. 32, 40 (2000). Respondent likewise concedes that "drug sniffs are designed, and if properly conducted are generally likely, to reveal only the presence of contraband." Brief for Respondent 17. Although respondent argues that the error rates, particularly the existence of false positives, call into question the premise that drug-detection dogs alert only to contraband, the record contains no evidence or findings that support his argument. Moreover, respondent does not suggest that an erroneous alert, in and of itself, reveals any legitimate private information, and, in this case, the trial judge found that the dog sniff was sufficiently reliable to establish probable cause to conduct a full-blown search of the trunk.

    Accordingly, the use of a well-trained narcotics-detection dog-one that "does not expose noncontraband items that otherwise would remain hidden from public view," Place, 462 U.S., at 707-during a lawful traffic stop, generally does not implicate legitimate privacy interests. In this case, the dog sniff was performed on the exterior of respondent's car while he was lawfully seized for a traffic violation. Any intrusion on respondent's privacy expectations does not rise to the level of a constitutionally cognizable infringement.

    This conclusion is entirely consistent with our recent decision that the use of a thermal-imaging device to detect the growth of marijuana in a home constituted an unlawful search. Kyllo v. United States, 533 U.S. 27 (2001). Critical to that decision was the fact that the device was capable of detecting lawful activity-in that case, intimate details in a home, such as "at what hour each night the lady of the house takes her daily sauna and bath." Id., at 38. The legitimate expectation that information about perfectly lawful activity will remain private is categorically distinguishable from respondent's hopes or expectations concerning the nondetection of contraband in the trunk of his car. A dog sniff conducted during a concededly lawful traffic stop that reveals no information other than the location of a substance that no individual has any right to possess does not violate the Fourth Amendment.

Aren't the handler's thoughts always essentially the same, always "wanting" the dogs to alert?

Since the handlers don't know which, if any, car they approach with the dogs have drugs, their thoughts, body language, etc., are the same for all the cars - ie. they're wanting/hoping the dog alerts?

iow, how/why were the dog handler's thoughts/body-language/whatever different when the dog's handlers approached this particular teacher's car compared to the dozens/hundreds of other cars the handlers are directed to approach that the dogs didn't alert to?

A trooper with the Iowa State Patrol pulled over 38-year-old Michael Spahlinger of Cleveland Heights, Ohio, on Thursday to perform a routine search. The trooper eventually called in a K-9 unit.

According to authorities, Spahlinger was driving a GMC pickup pulling a horse trailer on Interstate Highway 80 eastbound near Council Bluffs.

The K-9 unit helped uncover marijuana in a natural void in the front of the horse that was covered with hay bales and horse manure.